format: make comment directive regex more maintainable

Split over multiple lines, order entries alphabetically, avoid
unnecessary capturing groups for tiny performance gain.

Moves NOSONAR to inside the "correct" set of parenthesis, with no
functional effect though.

Author: scop

format/format.go

@@ -313,17 +313,26 @@ func (f *fumpter) lineEnd(line int) token.Pos {
 //
 //	//go:          | standard Go directives, like go:noinline
 //	//some-words:  | similar to the syntax above, like lint:ignore or go-sumtype:decl
-//	//line         | inserted line information for cmd/compile
 //	//export       | to mark cgo funcs for exporting
 //	//extern       | C function declarations for gccgo
-//	//sys(nb)?     | syscall function wrapper prototypes
-//	//nolint       | nolint directive for golangci
+//	//line         | inserted line information for cmd/compile
 //	//noinspection | noinspection directive for GoLand and friends
+//	//nolint       | nolint directive for golangci
 //	//NOSONAR      | NOSONAR directive for SonarQube
+//	//sys(nb)?     | syscall function wrapper prototypes
 //
 // Note that the "some-words:" matching expects a letter afterward, such as
 // "go:generate", to prevent matching false positives like "https://site".
-var rxCommentDirective = regexp.MustCompile(`^([a-z-]+:[a-z]+|line\b|export\b|extern\b|sys(nb)?\b|no(lint|inspection)\b)|NOSONAR\b`)
+var rxCommentDirective = regexp.MustCompile(
+	`^(?:` +
+		`[a-z-]+:[a-z]+` +
+		`|export` +
+		`|extern` +
+		`|line` +
+		`|no(?:inspection|lint)` +
+		`|NOSONAR` +
+		`|sys(?:nb)?` +
+		`)\b`)
 
 func (f *fumpter) applyPre(c *astutil.Cursor) {
 	f.splitLongLine(c)