Allow removal of max_output_tokens by setting GEN_AI_MAX_OUTPUT_TOKENS=0 (#1958)

Co-authored-by: Emerson Gomes <emerson.gomes@thalesgroup.com>

backend/danswer/llm/chat_llm.py

@@ -266,7 +266,7 @@ class DefaultMultiLLM(LLM):
                 stream=stream,
                 # model params
                 temperature=self._temperature,
-                max_tokens=self._max_output_tokens,
+                max_tokens=self._max_output_tokens if self._max_output_tokens > 0 else None,
                 timeout=self._timeout,
                 **self._model_kwargs,
             )