Validate Oracle dynamic column references in parser benchmarks

[cangelosilima]

The parser benchmark previously skipped column-reference validation for all dynamic SQL, allowing Oracle dynamic column lineage to regress without failing the benchmark.

Require the expected Id column for Oracle dynamic statements while retaining the existing SQL Server dynamic object-reference checks.

Validation:

• Both parser benchmark cases passed locally in Release.
• GitHub CI passed: CLI builds and tests on Windows and Linux, CLI and site coverage, performance budgets, CodeQL, and the Quality gate.
• Local site tests encountered two failures in unchanged code (localStorage availability and vendored model checksum/size). The local quality hook was bypassed after documenting those failures; the corresponding site checks passed in CI.

[github-actions]

Quality results

Quality gate: success
Commit: e2665e9ed6749032f4ade20deba49fde3d7c588f · View run and reports

Check	Result
Dependency vulnerabilities	success
CodeQL	success
CLI lint and format	success
CLI tests and coverage	success
CLI performance budgets	success
Site lint and format	success
Site tests and coverage	success
Workflow lint	success
Coverage badge data	success

Line coverage	Measured	Minimum
CLI	100%	100%
Site	100%	80%

CLI line, branch, method, and full method coverage must each reach 100%. A missing report is shown as unavailable. See the linked run for test failures, detailed CLI/Site coverage summaries and HTML reports, and scanner results.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-26T13:38:06.671413Z	e2665e9	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.