refactor: ZMQ context and proxy

Use ZMQ's global context instance and setup an XPUB/XSUB proxy intermediary to allow for easier multi-pubs.

close: N25B-217

src/control_backend/agents/transcription/transcription_agent.py

@@ -10,7 +10,6 @@ from spade.message import Message
 
 from control_backend.agents.transcription.speech_recognizer import SpeechRecognizer
 from control_backend.core.config import settings
-from control_backend.core.zmq_context import context as zmq_context
 
 logger = logging.getLogger(__name__)
 
@@ -47,7 +46,8 @@ class TranscriptionAgent(Agent):
             """Share a transcription to the other agents that depend on it."""
             receiver_jids = [
                 settings.agent_settings.text_belief_extractor_agent_name
-                + '@' + settings.agent_settings.host,
+                + "@"
+                + settings.agent_settings.host,
             ]  # Set message receivers here
 
             for receiver_jid in receiver_jids:
@@ -68,7 +68,7 @@ class TranscriptionAgent(Agent):
         return await super().stop()
 
     def _connect_audio_in_socket(self):
-        self.audio_in_socket = zmq_context.socket(zmq.SUB)
+        self.audio_in_socket = azmq.Context.instance().socket(zmq.SUB)
         self.audio_in_socket.setsockopt_string(zmq.SUBSCRIBE, "")
         self.audio_in_socket.connect(self.audio_in_address)