fix: handle XGo AST extension nodes consistently

[aofei]

Shared AST utilities, expression validation, and smart formatting had not kept pace with XGo-specific syntax nodes added over time. Matrix literal parsing fixes in #2755 exposed one of those gaps, where valid matrix nodes could reach traversal and formatting paths that did not handle them consistently.

Walk embedded expressions in domain text literals, class field tags, matrix literals, and element ellipses, and keep for-phrase traversal in source order. Preserve exported enum values in FileExports.

Allow matrix literals in parser expression validation and recurse through XGo expression, type, decorator, receiver, and tag nodes during smart formatting so imports are kept or removed based on actual use.

Updates #2755

[codecov]

Codecov Report

❌ Patch coverage is 96.00000% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 94.26%. Comparing base (d855433) to head (76bcb85).

Files with missing lines	Patch %	Lines
x/format/stmt_expr_or_type.go	95.23%	3 Missing and 2 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2806      +/-   ##
==========================================
+ Coverage   94.22%   94.26%   +0.04%     
==========================================
  Files          32       32              
  Lines       10454    10556     +102     
==========================================
+ Hits         9850     9951     +101     
+ Misses        431      430       -1     
- Partials      173      175       +2     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[gemini-code-assist]

Code Review

This pull request adds support for several new AST nodes and constructs (such as EnumType, MatrixLit, ElemEllipsis, DomainTextLitEx, and decorators) across the AST filtering, walking, parsing, and formatting packages, along with comprehensive unit tests. The review feedback focuses on adding defensive nil checks in the formatting code to prevent potential nil pointer dereferences when processing DomainTextLitEx, decorators, and EnumType specifications.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[fennoai]

Review summary

Solid, well-targeted change. The core fix is correct: ast.Walk now traverses the XGo-specific extension nodes (DomainTextLit args, ValueSpec.Tag, MatrixLit, ElemEllipsis), parser.checkExpr accepts MatrixLit, ast/filter.go filters EnumType specs, and x/format recurses through the matching node set so imports are kept/removed based on actual use. The ForPhrase reorder to source order in walk.go is consistent with formatForPhrase. Build and go test ./ast/ ./parser/ ./x/format/ pass, and the added unit tests cover the new traversal paths.

A few non-blocking notes below — one inline, and some pre-existing gaps surfaced by this consistency audit.

Pre-existing x/format coverage gaps (optional follow-ups)

This PR audits manually-maintained AST consumers, which surfaces a few spots x/format still does not cover. None are regressions from this PR, but they belong to the same class of "drop a used import" risk the PR is closing:

• x/format/gopstyle.go formatFile: the decl switch handles only *ast.FuncDecl and *ast.GenDecl; *ast.OverloadFuncDecl falls through silently. Its Funcs may contain *ast.SelectorExpr referencing an imported package (e.g. func F = (pkg.Method)), so that import would not be marked used.
• x/format/stmt_expr_or_type.go formatForStmt: skips v.Post, while ast.Walk visits it.
• x/format/stmt_expr_or_type.go formatFuncType: skips t.TypeParams, while ast.Walk visits them (e.g. func f[T fmt.Stringer]()).

Minor doc note (pre-existing, out of scope)

CLAUDE.md line 22 ("Printer shares test cases with parser - do NOT create separate test files in printer/_testdata/") is inaccurate: printer/_testdata/ exists with its own cases and TestFromTestdata runs them separately from TestFromParse. Not changed by this PR, but it sits right beside the lines you edited, so worth fixing while you're here.