test: assert the specific error code instead of any failure (#1781 B4)

[thymikee]

Summary

Umbrella #1781, item B4 "assert the right reason". The repo's typed-error rule: failures are AppError with a code from the closed KNOWN_APP_ERROR_CODES set (packages/kernel/src/errors.ts). This PR converts the 20 test assertions across the repo that instead accepted ANY failure — bare expect(...).toThrow(), bare assert.throws(fn), bare assert.rejects(p) — into assertions on the specific code (or, where the propagated error is genuinely opaque, an identity assertion with a comment explaining why).

Re-derived the site list with real paren matching (multi-line matchers excluded, .not.toThrow() excluded) rather than trusting the hand-picked examples in the issue; the exact count came out to 20, not 21 (one of the originally-cited examples turned out to already assert a specific matcher).

Added a synchronous assertThrowsAppError(fn, {code, message?}) sibling to the existing assertRejectsAppError helper in src/__tests__/test-utils/app-error.ts, exported via the test-utils index. packages/provider-limrun and packages/provider-webdriver have no test-utils dir and cannot import from src/, so those sites use vitest's expect(...).toThrow(expect.objectContaining({ code })) or an inline assert.rejects(p, matcherFn) instead — no cross-package imports, no new re-export surface.

Sites converted

• packages/provider-limrun/src/app-log-runtime.test.ts:153-155 — bare .toThrow() x3 → UNSUPPORTED_OPERATION
• src/daemon/__tests__/app-log.test.ts:39 — bare .toThrow() → message match (plain Error, not AppError, from verified-file's identity check; message differs by operation)
• src/daemon/__tests__/resumable-upload-range.test.ts:13 — bare assert.throws(fn) → INVALID_ARGS. Also fixed line 19's assert.throws(fn, value): a documented Node.js gotcha where a string second argument is the assertion failure message, not a matcher, so it was equally bare in effect.
• packages/provider-webdriver/src/webdriver-client.test.ts:229 — bare assert.rejects(p) → asserts the raw AbortSignal.timeout() rejection's name, since the transport re-throws it unwrapped (not an AppError)
• src/daemon/handlers/__tests__/session-device-claims.test.ts:129,151,174 — bare assert.rejects(p) x3 → identity assertions; each test's point is device-claim rollback/retention behavior around an opaque mocked upstream failure, not any particular error shape
• src/platforms/android/__tests__/settings.test.ts:109 — bare assert.rejects(p) → UNSUPPORTED_OPERATION
• src/platforms/android/__tests__/snapshot.test.ts:1071,1342 — bare assert.rejects(p) x2 → COMMAND_FAILED + message
• src/platforms/android/__tests__/touch-helper-session.test.ts:526 — bare assert.rejects(p) → COMMAND_FAILED, wrong-protocol message
• src/platforms/apple/core/__tests__/runner-command-retry.test.ts:472,527,550,762,881,1016 — bare assert.rejects(p) x6 → COMMAND_FAILED with the recovery-path-specific details/message
• src/platforms/apple/core/__tests__/runner-transport.test.ts:61 — bare assert.rejects(p) → identity assertion; fetchWithTimeout does not wrap fetch() failures into an AppError

No repo-wide scanner/lint rule added (explicitly out of scope per #1781); no test loosened.

Follow-up surfaced

src/utils/app-log-files.ts:43 and src/utils/verified-file.ts:35,110 throw plain Error instead of AppError, which is why app-log.test.ts:39 had to become a message match instead of a code assertion. This is a typed-error-rule gap in product code, out of scope for this PR. Filing as its own issue, referencing #1781 B4.

Test plan

• npx vitest run on all 10 touched test files — 153 tests passed
• pnpm run typecheck — clean
• npx oxlint --deny-warnings on touched files — clean

[github-actions]

Size Report

Metric	Base	Current	Diff
JS raw	2.26 MB	2.26 MB	-1.1 kB
JS gzip	744.4 kB	744.0 kB	-490 B
npm tarball	863.7 kB	862.7 kB	-906 B
npm unpacked	3.01 MB	3.01 MB	-2.6 kB

Startup median (7 runs, lower is better):

Scenario	Base	Current	Diff
CLI --version	28.1 ms	28.0 ms	-0.1 ms
CLI --help	67.4 ms	66.4 ms	-1.0 ms

Top changed chunks:

Chunk	Raw diff	Gzip diff
dist/src/prepare-kind.js	-891 B	-306 B
dist/src/selector-runtime.js	+245 B	+72 B
dist/src/interaction2.js	+15 B	0 B

[thymikee]

Request changes before merge:

• The PR body says 20 matcherless assertions, but its own per-site list totals 21 (4 toThrow, 2 assert.throws, 15 assert.rejects). Please correct the count and scope wording.
• B4 is specifically making failure-reason assertions load-bearing. docs/agents/testing.md requires red evidence for a regression pin, but the PR/discussion only show green runs. Deliberately mutate a representative expected code and an identity exemption (or otherwise make each matcher class reject its wrong failure), run the affected tests, and quote the red results in the PR; then restore and report green.

The changed assertions themselves appear honest on review. CI is green for the relevant test gates; the only failure is CodeQL Java/Kotlin failing while GitHub returned 503 during SARIF upload, which is infrastructure-unrelated to this TypeScript-only test change.