update README

feat: add /v1/responses api

fix: add timeout back to the correct position

fix: remove timeout

fix: do not response 200 if upstream is not success

fix: incorrect headers

fix: too little data for content-length

fix: proxy should forward headers and status

Add GET /v1/models and /v1/models/{model} endpoints for OpenAI API compatibility (#13)

remove experimental Anthropic endpoint (#11)

Fix refresh (#10)

[feature]: Anthropic endpoint (#9)

fix: do not streaming non-stream response (#8)

feat: Proxy embeddings API (#6)

fix: auto-release (#4)