refactor(gui): react-doctor cleanup for App, Logs, Subagents

[Wibias]

Summary

• App: client-resource polling + status-checked fetches.
• Logs/Subagents: fetch status checks, busy/loading guards, Logs tab keydown helper.
• Keeps current viewMode / SubagentsWorkspace integration.

Stack

• Depends on fix(gui): react-doctor shared foundations #466 (foundations).

Test plan

• bun x tsc -b in gui/
• Smoke App nav health/claude toggle, Logs tabs/refresh, Subagents save (incl. double-click)

Summary by CodeRabbit

• New Features

  • Logs and Debug tabs now support keyboard navigation and persist the selected tab in the URL hash.
  • Runtime version refreshes automatically, showing a fallback when health info is unavailable.

• Bug Fixes

  • Claude connection toggle is more reliable under rapid clicking, with optimistic UI updates and rollback on errors.
  • Subagent “Save” now blocks concurrent edits during in-flight requests, with improved handling of invalid payloads and server error messages.

• Tests

  • Added/expanded tests covering Claude toggle race conditions, Subagent busy behavior, and logs tab keyboard/hash logic.

[coderabbitai]

Warning

Review limit reached

@Wibias, 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: ASSERTIVE

Plan: Pro Plus

Run ID: 7627a4e3-365e-43f4-8375-9e4e7168e5fa

📥 Commits

Reviewing files that changed from the base of the PR and between 3c75294 and c9ba752.

📒 Files selected for processing (2)

• gui/src/pages/Subagents.tsx
• gui/tests/subagents-busy-race.test.tsx

📝 Walkthrough

Walkthrough

The GUI now uses client-resource polling for runtime and Claude state, shared helpers for Logs tab navigation, and guarded JSON/error handling with busy-state protection for Subagents saves.

Changes

GUI state and interaction updates

Layer / File(s)	Summary
App client-resource polling and Claude toggle gui/src/App.tsx, gui/tests/claude-toggle-race.test.tsx	Health and Claude state use keyed client resources; Claude toggles update optimistically, serialize requests, and revert on failure.
Shared Logs tab navigation gui/src/pages/logs-tab-keydown.ts, gui/src/pages/Logs.tsx, gui/tests/logs-tab-keydown.test.ts	Tab hash parsing, selection, keyboard handling, and focus management are centralized and tested.
Subagents request validation and save state gui/src/pages/Subagents.tsx, gui/tests/subagents-classic.test.tsx, gui/tests/subagents-busy-race.test.tsx	Model responses use stricter JSON parsing, saves block concurrent edits, error bodies are parsed, controls are disabled while saving, and server-applied state is tested.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• lidge-jun/opencodex#466: Introduces the client-resource and JSON parsing primitives used by this PR.

Suggested labels: bug

Suggested reviewers: ingwannu, lidge-jun

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title matches the main refactor scope and names the affected GUI areas.
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

---

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

[Wibias]

Rebased onto upstream/dev (dropped foundations stack; #466 already landed).

• gui/src/api.ts is no longer in this PR diff — Codex token-storage security thread should clear as outdated.
• Kept App client-resource polling, Logs logs-tab-keydown helper + silent refresh, Subagents busy/status-checked load/save; dropped deleted SubagentsWorkspace / viewMode wiring during conflict resolution.
• Follow-up: empty-body guard for readJsonOrThrow + Subagents test mock aligned to text()-based fetch helpers.

Tip: e6649563. Verified: bun x tsc -b in gui/; logs/subagents/client-resource tests green (17).

[Wibias]

CodeRabbit review fixes (7dad03ca)

Addressed the three open CodeRabbit findings on this PR:

1. gui/src/App.tsx — Serialize Claude toggle
   Guarded toggleClaude with an in-flight ref + pending state; Switch is disabled while the PUT is in flight so overlapping optimistic updates cannot race.

2. gui/src/pages/logs-tab-keydown.ts — Focused tests
   Added gui/tests/logs-tab-keydown.test.ts covering ArrowLeft/ArrowRight/Home/End, unrelated keys, hash parsing (readTabFromHash / selectLogsTab), and focus targets.

3. gui/src/pages/Subagents.tsx — Busy-state consistency
   Disabled reorder/remove/toggle controls while busy; handlers also no-op when busy; reconcile chosen from d.applied after a successful save.

Skipped the outdated Codex comment on gui/src/api.ts token storage (foundations; not in this PR).

Verification

• bun test tests/logs-tab-keydown.test.ts tests/subagents-classic.test.ts tests/subagents-classic.test.tsx — 11 pass
• bun x tsc -b — pass
• bun run lint (gui) — pass

[Wibias]

Verify / fix status (Claude toggle + Subagents busy races)

Baseline: origin tip was already 7dad03ca (CodeRabbit pass). Behavior was already complete — no production code changes.

Verified behavior (unchanged):

1. Claude toggle PUTs serialized via claudeToggleInFlight + switch disabled while pending (gui/src/App.tsx)
2. Subagents blocks toggle/remove/reorder while busy, then reconciles chosen from d.applied after successful save

Added tests (new SHA 3c75294b):

• gui/tests/claude-toggle-race.test.tsx — rapid-click issues one PUT until settle
• gui/tests/subagents-busy-race.test.tsx — edits blocked during save; applied reconciliation

Verification run:

• bun run typecheck (root) — pass
• bun x tsc -b --noEmit (gui) — pass
• bun test tests (gui, 176 tests) — pass
• bun run test (root suite) — pass

Result: behavior already complete at 7dad03ca; tests-only follow-up pushed as 3c75294b.