feat: support gemini openai api

[SkyeYoung]

Description

Refer to the changes in #12565 to add support for gemini

also check gemini openai doc: https://ai.google.dev/gemini-api/docs/openai

Already tested in real world:

# config.yaml
deployment:
  role: data_plane
  role_data_plane:
    config_provider: yaml

routes:
  - id: 1
    uri: /hello
    plugins:
      ai-proxy:
        provider: gemini
        auth:
          header:
            Authorization: "Bearer xxx"
        options:
          model: gemini-2.0-flash
#END

ubuntu-20:~/apisix$ curl -X POST http://localhost:9080/anything   -H "Content-Type: application/json"   -d '{"messages": [{"role": "user", "content": "Hello"}]}'
data: {"choices":[{"delta":{"content":"Hello! How can I help you today?","role":"assistant"},"finish_reason":"stop","index":0}],"created":1767957025,"id":"IeJgacrFI4OfqtsPuaDzuQ8","model":"gemini-2.5-flash","object":"chat.completion.chunk"}

data: [DONE]

Which issue(s) this PR fixes:

Fixes #

Checklist

• I have explained the need for this PR and the problem it solves
• I have explained the changes or the new features added to this PR
• I have added tests corresponding to this change
• I have updated the documentation to reflect this change
• I have verified that this change is backward compatible (If not, please discuss on the APISIX mailing list first)

[Copilot]

Pull request overview

This PR adds support for Google's Gemini OpenAI-compatible API to APISIX's AI proxy plugins. It follows the same implementation pattern as the recently added Azure OpenAI and AIMLAPI support.

Changes:

• Added a new driver for gemini-openai provider that uses Google's OpenAI-compatible endpoint
• Updated all plugin schemas to include gemini-openai in the provider enum lists
• Added comprehensive test coverage with a new test file
• Updated documentation in both English and Chinese to reflect the new provider option

Reviewed changes

Copilot reviewed 11 out of 11 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
apisix/plugins/ai-drivers/gemini-openai.lua	New driver implementation for Gemini's OpenAI-compatible API endpoint
apisix/plugins/ai-drivers/schema.lua	Added gemini-openai to provider schemas and compatibility checker
apisix/plugins/ai-proxy/schema.lua	Added gemini-openai to provider enums for both single and multi-instance schemas
apisix/plugins/ai-request-rewrite.lua	Added gemini-openai to provider enum list
docs/en/latest/plugins/ai-proxy.md	Added documentation for gemini-openai provider with endpoint details
docs/en/latest/plugins/ai-proxy-multi.md	Added documentation for gemini-openai provider with endpoint details
docs/en/latest/plugins/ai-request-rewrite.md	Added gemini-openai to provider options list
docs/zh/latest/plugins/ai-proxy.md	Chinese documentation update for gemini-openai provider
docs/zh/latest/plugins/ai-proxy-multi.md	Chinese documentation update for gemini-openai provider
docs/zh/latest/plugins/ai-request-rewrite.md	Chinese documentation update for gemini-openai provider
t/plugin/ai-proxy-gemini-openai.t	New test file with comprehensive test coverage for the gemini-openai provider

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Debug print statements should be removed before merging. These lines appear to be left over from development and testing.

Suggested change

	print "Hello, World!\n";
	print $resp;

[Copilot]

The endpoint configuration in TEST 3 uses port 7737 which differs from port 6724 used elsewhere in the test file. However, the mock server is only configured to listen on port 6724 (line 52). This will cause TEST 4 to fail as it tries to connect to a non-existent server on port 7737.

Suggested change

	"endpoint": "http://localhost:7737/v1/chat/completions"
	"endpoint": "http://localhost:6724/v1/chat/completions"

[Copilot]

Spelling error: "deekseek" should be "deepseek" to match the correct provider name used throughout the codebase.

[Copilot]

Spelling error: "deekseek" should be "deepseek" to match the correct provider name used throughout the codebase.