Add LLMRunFrame to trigger an LLM response, replacing context_aggregator.user().get_context_frame()

examples/foundational/22c-natural-conversation-mixed-llms.py

@@ -18,6 +18,7 @@ from pipecat.frames.frames import (
     Frame,
     FunctionCallInProgressFrame,
     FunctionCallResultFrame,
+    LLMRunFrame,
     StartFrame,
     StartInterruptionFrame,
     StopInterruptionFrame,
@@ -575,7 +576,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
                 "content": "Start by just saying \"Hello I'm ready.\" Don't say anything else.",
             }
         )
-        await task.queue_frames([context_aggregator.user().get_context_frame()])
+        await task.queue_frames([LLMRunFrame()])
 
     @transport.event_handler("on_app_message")
     async def on_app_message(transport, message):