feat(ai-proxy): support qiniu ai provider

[jiahuipaung]

Ⅰ. Describe what this PR did

Add Qiniu (七牛云) AI provider support to the ai-proxy plugin.

The Qiniu provider is OpenAI-compatible and routes requests to api.qnaigc.com. It supports chat completions and model listing APIs.

Files changed:

• provider/qiniu.go: New provider implementation
• provider/provider.go: Register qiniu provider type
• test/qiniu.go: Unit tests for the Qiniu provider
• main_test.go: Register TestQiniu test suite
• README.md: Add Qiniu provider documentation and usage example

Ⅱ. Does this pull request fix one issue?

N/A

Ⅲ. Why don't you add test cases (unit test/integration test)?

Test cases are included. All tests pass:

• RunQiniuParseConfigTests: config parsing validation
• RunQiniuOnHttpRequestHeadersTests: request header transformation (Host, Authorization)
• RunQiniuOnHttpRequestBodyTests: request body processing and model mapping

Ⅳ. Describe how to verify it

1. Run unit tests:

cd plugins/wasm-go/extensions/ai-proxy
go test -v -run TestQiniu ./...

2. Integration test with envoy.yaml (replace token with a valid Qiniu API token):

docker run --rm -it \
  -v $(pwd)/out/plugin.wasm:/etc/envoy/plugin.wasm \
  -v $(pwd)/envoy.yaml:/etc/envoy/envoy.yaml \
  -p 10000:10000 \
  --entrypoint /usr/local/bin/envoy \
  higress-registry.cn-hangzhou.cr.aliyuncs.com/higress/gateway:latest \
  -c /etc/envoy/envoy.yaml

curl -X POST http://localhost:10000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-4", "messages": [{"role": "user", "content": "hello"}], "stream": false}'

Ⅴ. Special notes for reviews

The Qiniu provider follows the same pattern as other OpenAI-compatible providers. No special handling is needed — it simply rewrites the Host header to api.qnaigc.com and injects the Bearer token.

Ⅵ. AI Coding Tool Usage Checklist (if applicable)

Please check all applicable items:

• For regular updates/changes (not new plugins):
  • [] I have provided the prompts/instructions I gave to the AI Coding tool below
  • [*] I have included the AI Coding summary below

AI Coding Prompts (for regular updates)

AI Coding Summary

Key decisions made:

• Followed the existing OpenAI-compatible provider pattern (same as groq, fireworks, deepseek)
• Used api.qnaigc.com as the provider domain
• Registered capabilities for ChatCompletion and Models APIs only, as Qiniu's API is OpenAI-compatible

Major changes implemented:

• New qiniuProvider struct implementing the Provider interface
• TransformRequestHeaders: rewrites Host to api.qnaigc.com, injects Bearer token, removes Content-Length
• ValidateConfig: requires at least one API token
• Registered providerTypeQiniu = "qiniu" in the provider initializers map

Important considerations:

• No custom request/response transformation needed; Qiniu's API is fully OpenAI-compatible