Always render audio from the current set of participants

We forgot to tell React that we need the audio renderer to react to changes in the set of MatrixRTC participants; instead we had it referencing rtcSession.memberships non-reactively.

Now, I'm not 100% confident that this is going to fix the "speaking from the void" issues observed in the wild, because I can't reproduce them and, in my testing, the InCallView component always seemed to be rendered redundantly when the MatrixRTC participants change, even though we hadn't explicitly stated that it needs to react. (This makes sense as we haven't memoized the component.) But it's worth a shot.

src/room/InCallView.tsx

@@ -104,6 +104,7 @@ import { ConnectionLostError } from "../utils/errors.ts";
 import { useTypedEventEmitter } from "../useEvents.ts";
 import { MatrixAudioRenderer } from "../livekit/MatrixAudioRenderer.tsx";
 import { muteAllAudio$ } from "../state/MuteAllAudioModel.ts";
+import { useMatrixRTCSessionMemberships } from "../useMatrixRTCSessionMemberships.ts";
 
 const canScreenshare = "getDisplayMedia" in (navigator.mediaDevices ?? {});
 
@@ -250,6 +251,7 @@ export const InCallView: FC<InCallViewProps> = ({
     useExperimentalToDeviceTransportSetting,
   );
   const encryptionSystem = useRoomEncryptionSystem(rtcSession.room.roomId);
+  const memberships = useMatrixRTCSessionMemberships(rtcSession);
 
   const showToDeviceEncryption = useMemo(
     () =>
@@ -723,10 +725,7 @@ export const InCallView: FC<InCallViewProps> = ({
           </Text>
         )
       }
-      <MatrixAudioRenderer
-        members={rtcSession.memberships}
-        muted={muteAllAudio}
-      />
+      <MatrixAudioRenderer members={memberships} muted={muteAllAudio} />
       {renderContent()}
       <CallEventAudioRenderer vm={vm} muted={muteAllAudio} />
       <ReactionsAudioRenderer vm={vm} muted={muteAllAudio} />