chore: clear Effect language service suggestions

[juliusmarminge]

Problem

vp run -r typecheck printed about 1,590 suggestion-level diagnostics from @effect/tsgo 0.41, mostly since the TypeScript 7 upgrade (#10663). Real issues get buried under the noise.

Fix

• schemaNumber off in tsconfig.base.json. 688 hits were the generated codex app-server schema and most of the other ~410 were JSON wire contracts, which cannot carry NaN or Infinity.
• Rewrote the remaining ~460 suggestions into the forms the rules suggest: succeedSome/succeedNone, asSome, ignore/ignoreCause, catchIf/catchCauseIf, yield* err, orElseSucceed, forEach, filterOrFail, fromOption, Effect.fn, typed decode*, and URL.canParse instead of try/catch. Interrupt-aware catchCause handlers became catchCauseIf, which re-fails the cause the same way the ternaries did.
• Per-line suppressions with a reason where the pattern is intentional: abortControllerInEffect (the OpenCode event pump, plus a file-level suppression for the Claude adapter tests), runEffectInsideEffect (the Claude SDK history callback), and raceFirstWithSleepToTimeout (three interval-versus-signal races that are not timeouts). There are 3 more where the rewrite can't typecheck; the reasons are inline.

Two traps for anyone doing the same thing:

• Effect.void does not satisfy X | undefined under exactOptionalPropertyTypes. Use Effect.undefined there.
• Effect.ignore passed as an Effect.fn trailing pipeable must be written Effect.ignore().

Worth a look when reviewing:

• The inverted predicate in ChromiumKeys.ts.
• FirefoxCookies.ts, where mapError moved into an Effect.fn pipeable. Most of that diff is re-indentation.
• The retry restructure in managedTunnelStartup.ts.

Verification

• vp run -r typecheck reports 0 diagnostics. Before: ~1,590 suggestions.
• vp test run on every changed test file plus the sibling tests of changed production files: about 3,700 tests pass across server, desktop, client-runtime, contracts, relay, shared, effect-acp, and effect-codex-app-server.
• vp lint on the changed files shows the same warnings before and after. vp fmt --check is clean.

Done by Claude Opus 5.5 in Claude Code.

🤖 Generated with Claude Code

---

Summary by CodeRabbit

• Bug Fixes
  • Improved recovery when managed tunnel registration encounters retryable errors, while non-retryable errors continue to surface.
  • Improved handling of malformed pull-request links so they fall back to a recognized repository host or an unknown-host label.
  • Refined error and cancellation handling across desktop, server, and connection workflows; interruptions are less likely to be treated as ordinary failures.

[macroscopeapp]

Approvability

Verdict: Not approved

Macroscope's review found this PR not approvable — This is a repository-wide Effect refactor across production and test code, with several runtime error-handling, retry, decoding, and concurrency paths touched. It also adds static-analysis suppressions, including a tsconfig-level disable, and modifies authentication code, so the scope warrants human review.

^{You can add or adjust custom eligibility rules. 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	13.5 KiB	13.5 KiB	+25 B (+0.2%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.1 KiB	7.1 KiB	−14 B (−0.2%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.4 KiB	6.5 KiB	+39 B (+0.6%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	56.2 KiB	56.3 KiB	+44 B (+0.1%)	66.4 KiB	✅
Codex	Live turn messages	9	10	+1 (+11.1%)	21	✅
Claude	Total thread wire	13.5 KiB	13.5 KiB	−31 B (−0.2%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.1 KiB	7.1 KiB	−3 B (−0.0%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.5 KiB	6.4 KiB	−28 B (−0.4%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	57.0 KiB	57.0 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	9	9	0 (0.0%)	21	✅

Baseline: 3e2370f · PR result: 5755bc2 · 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: 114.0 KiB
• Claude decoded thread snapshot: 114.7 KiB

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