Merge upstream/main into fork/dev (8 commits)

[omegent-app]

8 upstream commits, nine conflicts — all mobile, concentrated in the composer and outbox files this fork rewrote for its queue feature.

The notable one

pingdotgg#6543 "steer active turns by default" removes upstream's freshThreadBusy guard from the outbox drain, converging on what this fork has done all along: hand ownership to the server, which queues follow-ups during an active turn. The fork's own comment predicted exactly that ("unlike upstream, which waits on threadBusy"), so the fork's version is kept — it is now the same behaviour, plus a "deferred" drain outcome upstream lacks. Reporting an editor hold as success used to clear retry state and spin beginDispatch → finish → effect forever, and upstream's boolean would not typecheck against the DrainOutcome the rest of that function returns.

Other resolutions

• composerImages — the fork's iOS guard (requestMediaLibraryPermissionsAsync hard-crashes without a usage string) and upstream's Android foreground handoff (fix(mobile): prevent OTA update restart crashes pingdotgg/t3code#6324) are independent, so both are kept: catch and finally, so a picker failure still reaches the caller and the handoff still ends.
• SettingsRouteScreen — took upstream's runAppUpdateCheck and its new "ready" state, kept the fork's bundleLabel fallback (upstream has no bundle label and falls through to null, blanking that row here).
• ThreadComposer — dropped upstream's re-added queueCount prop; the fork already declares it from the previous sync.
• ThreadDetailScreen / ThreadRouteScreen / use-thread-composer-state / thread-outbox-model — fork-only queue props, state and a comment; kept.

What review caught — and what it says about the verification

Two adversarial reviews found the merge left mobile with eight typecheck errors. Worse, the workspace typecheck I had run reported clean: the vpr shim was not on PATH, so grepping its output for error TS counted an empty stream. vitest never typechecks, so the full suite stayed green the whole time. pnpm typecheck surfaces all eight.

• T3ComposerEditor — the merge took upstream's wrapped native view (fix(mobile): preserve keyboard suggestions while typing pingdotgg/t3code#6323) but none of what it calls: the expo-paste-input wrapper import, the useNativePaste import, and the handlePaste binding. Paste-to-attach on Android would have been dead on arrival.
• SettingsRouteScreen — upstream's helper emits an "Update ready" state the fork's local UpdateCheckState union lacks, so the setter did not match the callback and the new === "ready" branch compared non-overlapping types. Now uses the AppUpdateCheckState that was already imported; reportUpdateFailure went with the local helper it served.
• thread-outbox-model — upstream keeps threadBusy in the contract even though fix(mobile): steer active turns by default pingdotgg/t3code#6543 stopped gating on it, and its new tests still pass it. Accepted again as an explicitly unused field rather than editing upstream's tests, so their assertions keep compiling across future syncs.
• activeThreadBusy — fix(mobile): steer active turns by default pingdotgg/t3code#6543 removed it upstream; the fork's copy was required by ThreadComposer, never read there or in ThreadDetailScreen, and superseded by sendEntersQueue. Removed end to end rather than passing a value nothing uses.

Verification

All 8 upstream commits verified present by comparing each commit's added lines against the merged tree — 100% for every one.

pnpm typecheck 0 errors · pnpm test — 273 files, 0 failures · vp build in apps/web · vp check --fix.

Merge this with gh pr merge 404 --merge, not the button — the squash default is what cost us the lineage on #401. #403 adds a CI guard for exactly this.

Co-authored-by: Patrick Roza 42661+patroza@users.noreply.github.com

🤖 Generated with Claude Code via T3 Chat on Discord