fix: remove duplicate provider trace logging and dead code (#9278)

Provider traces were being created twice per step:
1. Via `request_async_with_telemetry` / `log_provider_trace_async` in LLMClient
2. Via direct `create_provider_trace_async` calls in LettaAgent

This caused duplicate records in provider_trace_metadata (Postgres) and
llm_traces (ClickHouse) for every agent step.

Changes:
- Remove redundant direct `create_provider_trace_async` calls from letta_agent.py
- Remove no-op `stream_async_with_telemetry` method (was just a pass-through to `stream_async`)
- Update callers to use `stream_async` directly

🤖 Generated with [Letta Code](https://letta.com)

Co-authored-by: Letta <noreply@letta.com>

letta/agents/letta_agent.py

@@ -1490,7 +1490,7 @@ class LettaAgent(BaseAgent):
                 # Attempt LLM request with telemetry wrapper
                 return (
                     request_data,
-                    await llm_client.stream_async_with_telemetry(request_data, agent_state.llm_config),
+                    await llm_client.stream_async(request_data, agent_state.llm_config),
                     current_in_context_messages,
                     new_in_context_messages,
                     valid_tool_names,

letta/llm_api/llm_client_base.py

@@ -127,14 +127,6 @@ class LLMClientBase:
                     except Exception as e:
                         logger.warning(f"Failed to log telemetry: {e}")
 
-    async def stream_async_with_telemetry(self, request_data: dict, llm_config: LLMConfig):
-        """Returns raw stream. Caller should log telemetry after processing via log_provider_trace_async().
-
-        Call set_telemetry_context() first to set agent_id, run_id, etc.
-        After consuming the stream, call log_provider_trace_async() with the response data.
-        """
-        return await self.stream_async(request_data, llm_config)
-
     async def log_provider_trace_async(
         self,
         request_data: dict,

letta/services/summarizer/summarizer.py

@@ -542,7 +542,7 @@ async def simple_summary(
             )
 
             # AnthropicClient.stream_async sets request_data["stream"] = True internally.
-            stream = await llm_client.stream_async_with_telemetry(req_data, summarizer_llm_config)
+            stream = await llm_client.stream_async(req_data, summarizer_llm_config)
             async for _chunk in interface.process(stream):
                 # We don't emit anything; we just want the fully-accumulated content.
                 pass