Remove chatml validation from sdk (#127)

* Removing chatml

* Adding prompt/completion tokens. Adding testing for streaming

* small fix to test file

* updated completion to return more

* removing dupes

agentops/langchain_callback_handler.py

@@ -433,6 +433,7 @@ async def on_llm_end(
             # TODO: more descriptive error
             error_event = ErrorEvent(trigger_event=self.events.llm[str(run_id)],
                                      details="on_llm_end: No generations", timestamp=get_ISO_time())
+                                     details="on_llm_end: No generations", timestamp=get_ISO_time())
             self.ao_client.record(error_event)
 
     @debug_print_function_params

agentops/llm_tracker.py

@@ -89,7 +89,7 @@ def generator():
             self.llm_event.agent_id = check_call_stack_for_agent_id()
             self.llm_event.prompt = kwargs["messages"]
             self.llm_event.prompt_tokens = response['usage']['prompt_tokens']
-            self.llm_event.completion = response['choices'][0]['message']['content']
+            self.llm_event.completion = {"role": "assistant", "content": response['choices'][0]['message']['content']}
             self.llm_event.completion_tokens = response['usage']['completion_tokens']
             self.llm_event.returns = {"content": response['choices'][0]['message']['content']}
             self.llm_event.model = response["model"]
@@ -181,7 +181,7 @@ async def async_generator():
             self.llm_event.agent_id = check_call_stack_for_agent_id()
             self.llm_event.prompt = kwargs["messages"]
             self.llm_event.prompt_tokens = response.usage.prompt_tokens
-            self.llm_event.completion = response.choices[0].message.model_dump().get('content')
+            self.llm_event.completion = response.choices[0].message.model_dump()
             self.llm_event.completion_tokens = response.usage.completion_tokens
             self.llm_event.returns = response.model_dump()
             self.llm_event.model = response.model