feat: implemented extra log level for LLM token stream

test/unit/agents/llm/test_llm_agent.py

@@ -49,6 +49,9 @@ async def test_llm_processing_success(mock_httpx_client, mock_settings):
     agent = LLMAgent("llm_agent")
     agent.send = AsyncMock()  # Mock the send method to verify replies
 
+    mock_logger = MagicMock()
+    agent.logger = mock_logger
+
     # Simulate receiving a message from BDI
     prompt = LLMPromptMessage(text="Hi", norms=[], goals=[])
     msg = InternalMessage(