* fix(vertex_llm_base.py): Handle api_base = ""
Fixes https://github.com/BerriAI/litellm/issues/5798
* fix(o1_transformation.py): handle stream_options not being supported
https://github.com/BerriAI/litellm/issues/5803
* docs(routing.md): fix docs
Closes https://github.com/BerriAI/litellm/issues/5808
* perf(internal_user_endpoints.py): reduce db calls for getting team_alias for a key
Use the list gotten earlier in `/user/info` endpoint
Reduces ui keys tab load time to 800ms (prev. 28s+)
* feat(proxy_server.py): support CONFIG_FILE_PATH as env var
Closes https://github.com/BerriAI/litellm/issues/5744
* feat(get_llm_provider_logic.py): add `litellm_proxy/` as a known openai-compatible route
simplifies calling litellm proxy
Reduces confusion when calling models on litellm proxy from litellm sdk
* docs(litellm_proxy.md): cleanup docs
* fix(internal_user_endpoints.py): fix pydantic obj
* test(test_key_generate_prisma.py): fix test