feat(lotus): io.lotus.set_audio_output — host-driven output device switch (#119)

The embed hides Element Call's footer, so its own output picker is out of
reach; the host's call bar now sends io.lotus.set_audio_output { deviceId }
and the fork applies it via mediaDevices.audioOutput.select(), answering
with io.lotus.audio_output_state { deviceId } on every change.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA

src/room/InCallView.tsx

@@ -36,6 +36,7 @@ import { startLotusQuality } from "../lotus/lotusQuality";
 import { startLotusDecorations } from "../lotus/lotusDecorations";
 import { startLotusDenoise } from "../lotus/lotusDenoise";
 import { startLotusCallSummary } from "../lotus/lotusCallSummary";
+import { startLotusAudioOutput } from "../lotus/lotusAudioOutput";
 import { startLotusDeafen } from "../lotus/lotusDeafen";
 import styles from "./InCallView.module.css";
 import { GridTile } from "../tile/GridTile";
@@ -315,6 +316,12 @@ export const InCallView: FC<InCallViewProps> = ({
   // audio (and optionally screenshare audio) at the LiveKit source. No-op
   // unless the host sends the action.
   useEffect(() => startLotusDeafen(), []);
+  // [lotus #119] Let the host pick the audio output device from its call bar.
+  const lotusMediaDevices = useMediaDevices();
+  useEffect(
+    () => startLotusAudioOutput(lotusMediaDevices),
+    [lotusMediaDevices],
+  );
 
   const fatalCallError = useBehavior(vm.fatalError$);
   // Stop the rendering and throw for the error boundary