ci: separate merge queues by category and independent scope

[tignear]

Problem and behavior

A lightweight title-check repair can inherit an earlier Veryl update in a serial speculative queue, as seen in queue PR #1322 (repair #1320 stacked on Veryl #1314). Split batching categories and use parallel file scopes so independent PR-policy repairs can advance separately.

• Lightweight PRs whose source CI passed lint and skipped both the Rust/NAPI aggregate and Ubuntu JS tests receive high priority. Running checks are not interrupted; overlapping running batches, barriers, and GitHub runner availability can still impose waiting.
• Queue rules: integration (Veryl, sync, release; one PR), dependencies (other Renovate; five PRs), changes (three PRs). Dependency/change batches collect for at most one minute. Head conditions are mutually exclusive.
• pr-policy is an exact allowlist of the title workflow and standalone title/release-impact/batch validator scripts and tests. Everything else, including unknown paths, shares core. A mixed change belongs to both scopes. Dependency updates and product features still share combined-tree validation because they interact.
• Parallel mode caps core at two simultaneous batches and PR policy at one, within the existing global ceiling of five. These are batch capacities, not runner quotas.
• Mergify configuration, release policy, and VERSION changes are barriers across scopes. Required checks, release hold/authorization, explicit merge requests, and zero automatic retries remain enforced. CI job selection is unchanged.

Validation

• 14 focused Node tests passed: real head routing, merge protections, independent title repairs, overlapping dependency/feature changes, mixed changes, unknown-path fallback, capacities, barriers, and lightweight priority conditions.
• Mergify CLI schema validation and git diff checks passed.
• Required pre-push checks run on push.

The default-branch Mergify configuration governs both master and develop. This takes effect after merge; existing queue runs were not cancelled or bypassed.

[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 2026-10-15T22:24:23.232511Z	4e92dbd	New commits

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.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4e92dbd770

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".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Prioritize docs using the check that actually runs

For a docs-only change, classifyFiles sets only docs, so .github/workflows/ci.yml runs Docs Build but skips Lint & Format; consequently this condition can never match even after all source CI passes, because Mergify distinguishes skipped checks from successful ones (condition reference). Such PRs therefore retain medium priority despite CONTRIBUTING.md explicitly stating that docs receive high priority. Add an alternative priority rule based on a successful Docs Build and the skipped product checks, or narrow the documented behavior.

Useful? React with 👍 / 👎.

[mergify]

Merge Queue Status

• ✅ Entered queue — 2026-10-15 23:14 UTC · Rule: default · triggered by rule Queue explicitly requested merges
• ✅ Checks passed · on draft merge queue: checking #1318 on master (b48dbb1), stacked on #1302 #1326
• ✅ Merged — 2026-10-16 00:02 UTC · at 3d7498bc91cbafc01e48a2c694e225f5e462ddc0 · merge

This pull request spent 48 minutes 37 seconds in the queue, including 46 minutes 12 seconds running CI.

Required conditions to merge

• check-success = Conventional Commit title
• check-success = Queue validation
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/Lint & Format
  • check-neutral = @github-actions/Lint & Format
  • check-skipped = @github-actions/Lint & Format
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/NAPI Build (Linux ARM64 GNU)
  • check-neutral = @github-actions/NAPI Build (Linux ARM64 GNU)
  • check-skipped = @github-actions/NAPI Build (Linux ARM64 GNU)
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/NAPI Build (Windows)
  • check-neutral = @github-actions/NAPI Build (Windows)
  • check-skipped = @github-actions/NAPI Build (Windows)
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/Rust Test & NAPI Build
  • check-neutral = @github-actions/Rust Test & NAPI Build
  • check-skipped = @github-actions/Rust Test & NAPI Build
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/JS Test (Ubuntu)
  • check-neutral = @github-actions/JS Test (Ubuntu)
  • check-skipped = @github-actions/JS Test (Ubuntu)
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/JS Test (Windows)
  • check-neutral = @github-actions/JS Test (Windows)
  • check-skipped = @github-actions/JS Test (Windows)
• any of [🛡 GitHub repository ruleset rule master]:
  • check-success = @github-actions/Conventional Commit title
  • check-neutral = @github-actions/Conventional Commit title
  • check-skipped = @github-actions/Conventional Commit title