feat(format): default output to dist/examples house style

Align the formatter defaults and layout with the hand-formatted reference
procedures in dist/examples so a clean `pgtidy fmt` produces the house style.

config.Default():
- align_param_types: false (no type-column alignment in param lists)
- plpgsql_declare_align_type / plpgsql_declare_align_eq: true

Formatter:
- routine header: leading-comma params at column 0, first param at one
  indent, RETURNS/LANGUAGE/volatility/SECURITY each indented one level
- %type / %rowtype printed tight (isPctTypeBoundary)
- DECLARE = / := / DEFAULT column padded only to the widest declaration
  that carries an assignment
- WHERE continuations in body UPDATE/DELETE: AND/OR aligned with WHERE
- EXCEPTION aligned to its enclosing BEGIN; column-0 comment continuations
  kept flush-left

Safety gate:
- SemanticallyEqual tolerates CRLF vs LF inside string literals (normNL);
  the formatter re-emits all layout with st.Newline, so a \r\n inside a
  multi-line string literal is normalisation, not a code change. This was
  why action_init and event_exec_func previously refused to format.

Corpus:
- add the four CRLF reference files as idempotence/safety fixtures
- regenerate test_a and test_mm_proc goldens

FOR...LOOP body indentation keeps the existing +1 convention (LOOP aligned
with FOR); the dist/examples use +2, so loop-body regions differ by
whitespace only.

pkg/format/body.go

@@ -177,7 +177,10 @@ func formatDeclareVars(b *strings.Builder, toks []cst.Tok, st config.Style) {
 			if nw > nameColW {
 				nameColW = nw
 			}
-			if tw > typeColW {
+			// typeColW drives the '='/':='/DEFAULT column (align_eq only), so
+			// only declarations that actually carry an assignment participate —
+			// a bare "name type;" must not widen it.
+			if tw > typeColW && declHasAssignment(body) {
 				typeColW = tw
 			}
 		}
@@ -207,7 +210,7 @@ func formatDeclareVars(b *strings.Builder, toks []cst.Tok, st config.Style) {
 			writeDeclareAligned(b, body, st, nameColW, typeColW)
 		} else {
 			for j, t := range body {
-				if j > 0 && needSpace(body[j-1].Tok, t.Tok) {
+				if j > 0 && needSpace(body[j-1].Tok, t.Tok) && !isPctTypeBoundary(body, j) {
 					b.WriteByte(' ')
 				}
 				b.WriteString(caseText(t.Tok, st))
@@ -220,6 +223,32 @@ func formatDeclareVars(b *strings.Builder, toks []cst.Tok, st config.Style) {
 	}
 }
 
+// declHasAssignment reports whether a DECLARE variable body (name type … ) has
+// a default assignment ( := / = / DEFAULT ) at paren depth 0.
+func declHasAssignment(body []cst.Tok) bool {
+	depth := 0
+	for _, t := range body {
+		switch t.Tok.Kind {
+		case lexer.LParen, lexer.LBracket:
+			depth++
+		case lexer.RParen, lexer.RBracket:
+			if depth > 0 {
+				depth--
+			}
+		}
+		if depth != 0 {
+			continue
+		}
+		if t.Tok.Kind == lexer.Operator && (t.Tok.Text == ":=" || t.Tok.Text == "=") {
+			return true
+		}
+		if t.Tok.Kind == lexer.Ident && lowerASCII(t.Tok.Text) == "default" {
+			return true
+		}
+	}
+	return false
+}
+
 // declareNameTypeWidth returns the rendered width of the name and type portions
 // of a DECLARE variable declaration (without the default assignment).
 // Format is: [name type [:= default]] or [name type [DEFAULT default]].
@@ -256,7 +285,7 @@ func declareNameTypeWidth(body []cst.Tok, st config.Style) (nameW, typeW int) {
 	}
 	var tb strings.Builder
 	for j, t := range typeTokens {
-		if j > 0 && needSpace(typeTokens[j-1].Tok, t.Tok) {
+		if j > 0 && needSpace(typeTokens[j-1].Tok, t.Tok) && !isPctTypeBoundary(typeTokens, j) {
 			tb.WriteByte(' ')
 		}
 		tb.WriteString(caseText(t.Tok, st))
@@ -315,7 +344,7 @@ func writeDeclareAligned(b *strings.Builder, body []cst.Tok, st config.Style, na
 
 	var typeStr strings.Builder
 	for j, t := range typeTokens {
-		if j > 0 && needSpace(typeTokens[j-1].Tok, t.Tok) {
+		if j > 0 && needSpace(typeTokens[j-1].Tok, t.Tok) && !isPctTypeBoundary(typeTokens, j) {
 			typeStr.WriteByte(' ')
 		}
 		typeStr.WriteString(caseText(t.Tok, st))
@@ -429,7 +458,12 @@ func formatBodyStatements(text string, st config.Style) string {
 			}
 		case "exception":
 			inException = true
-			effectiveDepth = 0
+			// EXCEPTION belongs to its nearest enclosing BEGIN, so align it one
+			// level in from the current block body (col 0 for the outermost).
+			effectiveDepth = blockDepth - 1
+			if effectiveDepth < 0 {
+				effectiveDepth = 0
+			}
 		}
 
 		baseIndent := strings.Repeat(st.Indent, effectiveDepth)
@@ -589,7 +623,13 @@ func formatBodyStmtLines(lines []bline, baseIndent string, st config.Style) []st
 	for i, ll := range lines {
 		text := ll.text
 		indent := baseIndent
-		if i > 0 && ll.indent != "" && !isStandaloneBodyKeyword(ll.text, "then", "else", "elsif", "elseif") {
+		switch {
+		case i > 0 && ll.indent == "" && len(significantBodyTokens(text)) == 0:
+			// A column-0 comment line trailing a multi-line (commented-out)
+			// statement is a continuation the author left flush-left — keep it
+			// there rather than re-indenting it to block depth.
+			indent = ""
+		case i > 0 && ll.indent != "" && !isStandaloneBodyKeyword(ll.text, "then", "else", "elsif", "elseif"):
 			indent = ll.indent
 		}
 		out = append(out, indent+text)
@@ -639,7 +679,9 @@ func reindentBodyDML(lines []bline, baseIndent string, st config.Style) []string
 			continue
 		}
 		if lineDepth == 0 && (kw == "and" || kw == "or") {
-			out = append(out, baseIndent+st.Indent+strings.TrimSpace(text))
+			// House style: AND/OR line up with the WHERE keyword; the first
+			// predicate is indented two levels under it.
+			out = append(out, baseIndent+strings.TrimSpace(text))
 			afterWhere = false
 			updateBodyParenDepth(text, &parenDepth)
 			continue

pkg/format/dml.go

@@ -462,7 +462,7 @@ func dmlInline(toks []cst.Tok, st config.Style) string {
 	}
 	var b strings.Builder
 	for i, t := range toks {
-		if i > 0 && needSpace(toks[i-1].Tok, t.Tok) {
+		if i > 0 && needSpace(toks[i-1].Tok, t.Tok) && !isPctTypeBoundary(toks, i) {
 			b.WriteByte(' ')
 		}
 		// Space after comma in calls: func(a, b) vs func(a,b).

pkg/format/format.go

@@ -147,8 +147,9 @@ func (p *printer) writeCreateFunction(cf *cst.CreateFunction) {
 		paramTexts = alignParamTypes(paramTexts)
 	}
 
-	first := p.st.Indent + " " // align item text one column past the comma
-	cont := p.st.Indent
+	// House style: first parameter indented one level; leading-comma
+	// continuation lines carry the comma at column 0 followed by one space.
+	first := p.st.Indent
 	for i, text := range paramTexts {
 		param := cf.Params[i]
 		p.nl()
@@ -159,8 +160,7 @@ func (p *printer) writeCreateFunction(cf *cst.CreateFunction) {
 				p.b.WriteString(",")
 			}
 		} else {
-			p.b.WriteString(cont)
-			p.b.WriteString(",")
+			p.b.WriteString(", ")
 			p.b.WriteString(text)
 		}
 		// Emit trailing inline comment from the separator (e.g. --description after param).
@@ -179,6 +179,7 @@ func (p *printer) writeCreateFunction(cf *cst.CreateFunction) {
 
 	for _, clause := range cf.Options {
 		p.nl()
+		p.b.WriteString(p.st.Indent)
 		p.b.WriteString(p.inline(clause))
 	}
 	if cf.As != nil {
@@ -218,7 +219,7 @@ func (p *printer) inline(toks []cst.Tok) string {
 	}
 	var b strings.Builder
 	for i, t := range toks {
-		if i > 0 && needSpace(toks[i-1].Tok, t.Tok) {
+		if i > 0 && needSpace(toks[i-1].Tok, t.Tok) && !isPctTypeBoundary(toks, i) {
 			b.WriteByte(' ')
 		}
 		var prev lexer.Token
@@ -254,6 +255,31 @@ func (p *printer) trailingComments(lead cst.Trivia) {
 // tightOps are operators printed without surrounding spaces.
 var tightOps = map[string]bool{"::": true, ":": true, "->": true, "->>": true}
 
+// isPctTypeBoundary reports whether the gap between toks[i-1] and toks[i] sits
+// inside a %TYPE / %ROWTYPE modifier (e.g. core.tbl%rowtype), which is printed
+// tight like "::" rather than as the modulo operator.
+func isPctTypeBoundary(toks []cst.Tok, i int) bool {
+	if i <= 0 || i >= len(toks) {
+		return false
+	}
+	isPct := func(t lexer.Token) bool { return t.Kind == lexer.Operator && t.Text == "%" }
+	isTypeWord := func(t lexer.Token) bool {
+		if t.Kind != lexer.Ident {
+			return false
+		}
+		l := lowerASCII(t.Text)
+		return l == "type" || l == "rowtype"
+	}
+	prev, cur := toks[i-1].Tok, toks[i].Tok
+	if isPct(prev) && isTypeWord(cur) {
+		return true // space after %
+	}
+	if isPct(cur) && i+1 < len(toks) && isTypeWord(toks[i+1].Tok) {
+		return true // space before %
+	}
+	return false
+}
+
 // parenKws are keywords that always take a space before '(' because they
 // introduce a subquery or a bracketed clause, not a function-call argument list.
 var parenKws = map[string]bool{

pkg/format/format_test.go

@@ -23,13 +23,13 @@ func TestFormatHeaderGolden(t *testing.T) {
 
 	want := "--select * from dropall('resolvespec_login');\n" +
 		"CREATE OR REPLACE FUNCTION resolvespec_login(\n" +
-		"   INOUT p_data    jsonb\n" +
-		"  ,OUT p_success boolean\n" +
-		"  ,OUT p_error   text\n" +
+		"  INOUT p_data jsonb\n" +
+		", OUT p_success boolean\n" +
+		", OUT p_error text\n" +
 		")\n" +
-		"LANGUAGE plpgsql\n" +
-		"VOLATILE\n" +
-		"SECURITY DEFINER\n" +
+		"  LANGUAGE plpgsql\n" +
+		"  VOLATILE\n" +
+		"  SECURITY DEFINER\n" +
 		"AS\n" +
 		"$$\nbegin end;\n$$;\n"
 
@@ -130,8 +130,8 @@ func TestFormatIssue1PLpgSQLIndenting(t *testing.T) {
 
 	want := "CREATE FUNCTION f(\n" +
 		")\n" +
-		"RETURNS void\n" +
-		"LANGUAGE plpgsql\n" +
+		"  RETURNS void\n" +
+		"  LANGUAGE plpgsql\n" +
 		"AS\n" +
 		"$$\n" +
 		"DECLARE\n" +
@@ -144,14 +144,14 @@ func TestFormatIssue1PLpgSQLIndenting(t *testing.T) {
 		"    set status = 'done'\n" +
 		"    where\n" +
 		"        u.rid_process = r_lp.rid_process\n" +
-		"      and nv(u.status) <> 'done';\n" +
+		"    and nv(u.status) <> 'done';\n" +
 		"  elsif r_lp.total > 0\n" +
 		"  then\n" +
 		"    update core.process u\n" +
 		"    set status = 'open'\n" +
 		"    where\n" +
 		"        u.rid_process = r_lp.rid_process\n" +
-		"      and nv(u.status) <> 'open';\n" +
+		"    and nv(u.status) <> 'open';\n" +
 		"\n" +
 		"  end if;\n" +
 		"$$;\n"

pkg/format/safety.go

@@ -32,6 +32,14 @@ func SemanticallyEqual(a, b string) bool {
 			if !strings.EqualFold(ta[i].Text, tb[i].Text) {
 				return false
 			}
+		case lexer.String, lexer.EscapeString, lexer.BitString, lexer.HexString, lexer.UnicodeString:
+			// A CRLF vs LF difference inside a multi-line string literal is a
+			// line-ending normalisation, not a change of code content — the
+			// formatter always re-emits layout with st.Newline. Compare the
+			// literal modulo \r\n ↔ \n.
+			if normNL(ta[i].Text) != normNL(tb[i].Text) {
+				return false
+			}
 		case lexer.DollarString:
 			_, innerA, _, okA := splitDollarQuote(ta[i].Text)
 			_, innerB, _, okB := splitDollarQuote(tb[i].Text)
@@ -47,6 +55,10 @@ func SemanticallyEqual(a, b string) bool {
 	return true
 }
 
+// normNL collapses CRLF to LF so string literals compare independent of the
+// source file's line-ending convention.
+func normNL(s string) string { return strings.ReplaceAll(s, "\r\n", "\n") }
+
 // significantTokens lexes src and returns its tokens excluding EOF and trivia
 // (whitespace/comments).
 func significantTokens(src string) []lexer.Token {