fix: voice_mode break fix (#8014)

* 🔧 (use-get-messages-polling.ts): refactor enqueuePolling method to simplify logic and improve readability
🔧 (use-get-messages-polling.ts): refactor startNextPolling method to remove unnecessary code and improve efficiency
🔧 (use-get-messages-polling.ts): refactor removeFromQueue method to simplify and improve maintainability
🔧 (use-get-messages-polling.ts): refactor useGetMessagesPollingMutation to handle stopping polling and removing from queue
🔧 (use-get-transactions.ts): add early return to getTransactionsFn to handle empty id case
🔧 (use-get-voice-list.ts): refactor useGetVoiceList function to handle empty elevenlabsApiKey case and improve readability
🔧 (audio-settings-dialog.tsx): update useGetVoiceList call to pass elevenLabsApiKey as argument

* queues should hold payloads not strings

* ✨ (use-get-global-variables.ts): add setGlobalVariablesEntities function to store global variables entities for use in the application
✨ (audio-settings-dialog.tsx): introduce debounced function to update ElevenLabs API key in global variables entities
📝 (globalVariables.ts): add setGlobalVariablesEntities function to the global variables store for managing entities
📝 (index.ts): define GlobalVariable type for global variables entities in the store

* ✨ (frontend): add support for extracting flowId from URL query parameters to allow direct linking to specific flow logs.

---------

Co-authored-by: cristhianzl <cristhian.lousa@gmail.com>

src/backend/base/langflow/api/v1/voice_mode.py

@@ -1143,8 +1143,8 @@ async def flow_tts_websocket(
     try:
         await client_websocket.accept()
 
-        openai_send_q: asyncio.Queue[str] = asyncio.Queue()
-        client_send_q: asyncio.Queue[str] = asyncio.Queue()
+        openai_send_q: asyncio.Queue[dict] = asyncio.Queue()
+        client_send_q: asyncio.Queue[dict] = asyncio.Queue()
 
         log_event = create_event_logger()
 
@@ -1171,13 +1171,13 @@ async def flow_tts_websocket(
         def openai_send(payload):
             log_event(payload, LF_TO_OPENAI)
             logger.trace(f"Sending text {LF_TO_OPENAI}: {payload['type']}")
-            openai_send_q.put_nowait(json.dumps(payload))
+            openai_send_q.put_nowait(payload)
             logger.trace("JSON sent.")
 
         def client_send(payload):
             log_event(payload, LF_TO_CLIENT)
             logger.trace(f"Sending JSON {LF_TO_CLIENT}: {payload['type']}")
-            client_send_q.put_nowait(json.dumps(payload))
+            client_send_q.put_nowait(payload)
             logger.trace("JSON sent.")
 
         async def close():