chore: Fail CI on high-confidence C# dead code and remove leftover readiness probes

[hatayama]

Summary

• Pull requests that touch package C# (or the scanner itself) now fail CI when high-confidence dead code is introduced.
• Three leftover execute-dynamic-code readiness probe helpers with zero callers are removed.

User Impact

• Before: the dead-code scanner existed but never ran in CI, so unused C# symbols could land unnoticed.
• After: Unused / UnusedPrivateMember / UnusedLocal findings fail the new Dead Code Gate. PublicCandidate and TestOnly still require manual review and do not fail CI.

Changes

• Add .github/workflows/dead-code.yml (scanner unit tests + --fail-on high-confidence scan; no advisory artifacts).
• Remove:
  • DynamicCodeForegroundWarmupRunner.TryRunBackgroundSequenceAsync — never had a caller since introduction.
  • FirstPartyToolsEditorStartup.CreateExecuteDynamicCodeReadinessProbeCode — leftover after commit 7e45f1e7 intentionally switched the Editor readiness probe from execute-dynamic-code to get-version so user-disabled tools cannot block startup. Rewiring the old path would reverse that design; deletion is correct.
  • ExecuteDynamicCodeReadinessProbe.CreatePrimaryReturnStringProbeCode — only called by the deleted helper above; removed in the same change.
• Document the CI gate in docs/dead-code-scanner.md.

Out of scope

• Issue Gate dead code in CI with a deliberate-keep attribute, and clear the current findings #1987 task 1 ([UnityCliLoopKeep(reason)]) is intentionally not included. There are no remaining intentional-keep call sites after this cleanup, public API shape cannot be renamed later, and a keep attribute can hide stale symbols instead of forcing review. Follow-up triage of PublicCandidate findings will decide whether any keep mechanism is needed.

Verification

• scripts/check-dead-code.sh ... --fail-on high-confidence → exit 0
• Before/after scan counts on origin/v3-beta: 244 → 242 (PublicCandidate 187 → 185)
• dotnet test tests/UnityCliLoop.DeadCodeScanner.Tests/... → Passed 4 / Failed 0
• cd cli/release-automation && go test ./internal/architecture -run 'TestWorkflowActions|TestPullRequestWorkflow' -count=1 → ok
• dist/darwin-arm64/uloop compile → ErrorCount 0, WarningCount 0

Refs #1987

[coderabbitai]

Warning

Review limit reached

@hatayama, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 41 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 91ae9ac1-7021-4509-a606-f810e7360518

📥 Commits

Reviewing files that changed from the base of the PR and between fa77edc and f29589f.

📒 Files selected for processing (4)

• .github/workflows/dead-code.yml
• docs/dead-code-scanner.md
• tests/UnityCliLoop.DeadCodeScanner.Tests/CommandLineOptionsTests.cs
• tests/UnityCliLoop.DeadCodeScanner.Tests/DeadCodeScannerTests.cs

📝 Walkthrough

Walkthrough

The pull request removes unused dynamic-code warmup and readiness helpers, adds a GitHub Actions dead-code scanning workflow for pull requests and manual runs, and documents the CI failure criteria.

Changes

Dead code cleanup and CI gate

Layer / File(s)	Summary
Remove obsolete dynamic-code paths Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/DynamicCodeForegroundWarmupRunner.cs, Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/ExecuteDynamicCodeReadinessProbe.cs, Packages/src/Editor/FirstPartyTools/FirstPartyToolsEditorStartup.cs	Removes the background warmup path and two public readiness-probe code-generation helpers.
Add dead-code CI validation .github/workflows/dead-code.yml, docs/dead-code-scanner.md	Adds automated scanner tests and high-confidence dead-code checks for selected pull-request changes, with documentation of failing and non-failing finding categories.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related issues

• hatayama/unity-cli-loop#1987 — Covers the dead-code CI gate, documentation, and removal of the obsolete dynamic-code helpers.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately summarizes the CI dead-code gate and the removal of leftover readiness probe helpers.
Description check	✅ Passed	The description matches the changeset and explains the CI gate, code removals, documentation update, and verification.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/dead-code-ci-gate

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 2

🧹 Nitpick comments (1)

.github/workflows/dead-code.yml (1)

3-11: 🩺 Stability & Availability | 🔵 Trivial

Confirm this path-filtered workflow is not a globally required check.

PRs targeting main or v3-beta that touch unrelated files will not receive a workflow result. If branch protection requires Dead Code Gate for every PR, those changes can be blocked; use an always-running workflow with conditional expensive steps instead.

Based on learnings, always-required checks should use an always-running workflow rather than a path-filtered workflow.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In @.github/workflows/dead-code.yml around lines 3 - 11, Update the workflow
trigger so it runs for every pull request targeting main or v3-beta, removing
the path filter from the on.pull_request configuration. Keep the dead-code scope
restrictions within conditional job or step execution so unrelated changes still
produce the required Dead Code Gate result without running expensive scanning
unnecessarily.

Source: Learnings

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In @.github/workflows/dead-code.yml:
- Around line 23-24: Add persist-credentials: false to the actions/checkout step
in the dead-code workflow, keeping the existing pinned checkout action
unchanged.

In `@docs/dead-code-scanner.md`:
- Around line 22-24: Update the workflow trigger description in the dead-code
documentation to state that it runs only for pull requests targeting main or
v3-beta, and include .github/workflows/dead-code.yml itself among the triggering
path changes. Preserve the existing listed source, scanner, test, and script
paths.

---

Nitpick comments:
In @.github/workflows/dead-code.yml:
- Around line 3-11: Update the workflow trigger so it runs for every pull
request targeting main or v3-beta, removing the path filter from the
on.pull_request configuration. Keep the dead-code scope restrictions within
conditional job or step execution so unrelated changes still produce the
required Dead Code Gate result without running expensive scanning unnecessarily.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 9faa9984-dbf8-4af5-8314-150b82ae7ea6

📥 Commits

Reviewing files that changed from the base of the PR and between ffc2737 and fa77edc.

📒 Files selected for processing (5)

• .github/workflows/dead-code.yml
• Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/DynamicCodeForegroundWarmupRunner.cs
• Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/ExecuteDynamicCodeReadinessProbe.cs
• Packages/src/Editor/FirstPartyTools/FirstPartyToolsEditorStartup.cs
• docs/dead-code-scanner.md

💤 Files with no reviewable changes (3)

• Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/ExecuteDynamicCodeReadinessProbe.cs
• Packages/src/Editor/FirstPartyTools/ExecuteDynamicCode/Execution/DynamicCodeForegroundWarmupRunner.cs
• Packages/src/Editor/FirstPartyTools/FirstPartyToolsEditorStartup.cs