fix: fix test race condition

ref: N25B-301

src/control_backend/agents/perception/transcription_agent/transcription_agent.py

@@ -37,7 +37,7 @@ class TranscriptionAgent(BaseAgent):
         self.speech_recognizer.load_model()  # Warmup
 
         # Start background loop
-        await self.add_behavior(self._transcribing_loop())
+        self.add_behavior(self._transcribing_loop())
 
         self.logger.info("Finished setting up %s", self.name)
 

src/control_backend/agents/perception/vad_agent.py

@@ -93,7 +93,7 @@ class VADAgent(BaseAgent):
         # Warmup/reset
         await self.reset_stream()
 
-        await self.add_behavior(self._streaming_loop())
+        self.add_behavior(self._streaming_loop())
 
         # Start agents dependent on the output audio fragments here
         transcriber = TranscriptionAgent(audio_out_address)