test: remove redundant and stale tests

[t3-code]

problem

a broad repository sweep found tests that no longer add regression signal because they duplicate canonical coverage or exercise helpers with no production callers.

change

• removed 3 duplicate tests:
  • one duplicate orchestration recovery scenario
  • two mobile nextTerminalId cases already covered in packages/shared
  • one desktop Tailnet IPv4 case already covered in packages/tailscale
• removed 26 stale tests for unreachable helpers:
  • web history bootstrap, terminal UI cleanup, orchestration event effects, and recovery helpers
  • desktop preview label-position helper
  • server non-negative integer invariant and abandoned provider snapshot helpers
  • relay URL-only env reconciliation helper
  • mobile showcase scene URL helper

evidence

all stale helpers were searched repository-wide excluding tests. each only has its definition remaining. git history confirms their production call sites were removed or replaced. the provider registry's current integration test remains and exercises the live persistence path.

sweep

three parallel hermes subagents inspected web/desktop, server/infra/scripts/contracts, and mobile/shared packages. their inventories covered 797+ tracked test files and over 5,000 scoped test declarations. independent searches rejected similar tests when they exercised distinct implementations, providers, or edge cases.

verification

• focused retained suites: 8 files passed, 110 tests passed
• web, mobile, server, desktop, relay, and scripts typechecks passed
• targeted lint passed
• targeted format passed
• git diff --check passed

request provenance

• requested by: @maria-rcks (maria on Discord)
• source: Discord request

implemented by gpt-5.6-sol via hermes agent (openai-codex).

---

Note

Low Risk
Test-only deletions with no production logic changes. Main risk is slightly reduced regression signal if any removed cases still covered unique behavior.

Overview
Removes redundant and stale tests across desktop, mobile, server, web, relay, and scripts. No production code changes.

Drops duplicate coverage already owned elsewhere (Tailnet IPv4 checks in packages/tailscale, mobile nextTerminalId cases in shared terminal helpers) plus unit tests for helpers that no longer have production callers, including web history/orchestration helpers, desktop label positioning, abandoned provider snapshot helpers, and a few deploy/showcase utilities.

^{Reviewed by Cursor Bugbot for commit 035cd44. Bugbot is set up for automated code reviews on this repo. Configure here.}

Note

Remove redundant and stale tests across desktop, mobile, server, and web

Deletes several obsolete test files and removes individual test cases that cover functions no longer being tested or that have drifted out of sync with the implementation.

• Deletes full test files: PickPreload.test.ts, historyBootstrap.test.ts, terminalUiStateCleanup.test.ts, orchestrationEventEffects.test.ts, and orchestrationRecovery.test.ts
• Removes individual test cases for isTailscaleIpv4Address, nextTerminalId, requireNonNegativeInteger, mergeProviderSnapshots/selectProvidersByKind, reconcileRootEnvRelayUrl, and showcaseSceneUrl from their respective test files

^{Macroscope summarized 035cd44.}

[macroscopeapp]

Approvability

Verdict: Approved 035cd44

This PR only removes test code from .test.ts files with no changes to production code. Test-only cleanup has no runtime impact.

^{You can customize Macroscope's approvability policy. Learn more.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	11.3 KiB	11.3 KiB	−24 B (−0.2%)	15.1 KiB	✅
Codex	Thread snapshot wire	5.5 KiB	5.5 KiB	−4 B (−0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	5.9 KiB	5.9 KiB	−20 B (−0.3%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	49.7 KiB	49.7 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	16	16	0 (0.0%)	21	✅
Claude	Total thread wire	11.3 KiB	11.3 KiB	−33 B (−0.3%)	15.1 KiB	✅
Claude	Thread snapshot wire	5.5 KiB	5.5 KiB	−11 B (−0.2%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	5.9 KiB	5.8 KiB	−22 B (−0.4%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	50.6 KiB	50.6 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	16	16	0 (0.0%)	21	✅

Baseline: 5a84614 · PR result: 035cd44 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 94.6 KiB
• Claude decoded thread snapshot: 95.4 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

Dismissing prior approval to re-evaluate 035cd44