feat: support vertex-ai

[SkyeYoung]

Description

Refer to the changes in #12565 to add support for vertex ai.

Already tested in real world:

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

routes:
  - id: 1
    uri: /anything
    plugins:
      ai-proxy:
        ssl_verify: false
        provider: vertex-ai
        provider_conf:
          project_id: evident-xxx
          region: us-central1
        auth:
          gcp:
            service_account_json: |
              {
                ...
              }
        options:
          model: google/gemini-2.5-flash
#END

❯ curl -X POST http://localhost:9080/anything -H "Content-Type: application/json" -d '{"messages": [{"role": "user", "content": "Hello"}]}'
{"model":"google/gemini-2.5-flash","id":"IthxaZicNsq4qsMPjJeJqAI","object":"chat.completion","usage":{"completion_tokens_details":{"reasoning_tokens":17},"extra_properties":{"google":{"traffic_type":"ON_DEMAND"}},"total_tokens":28,"prompt_tokens":1,"completion_tokens":10},"created":1769068578,"system_fingerprint":"","choices":[{"index":0,"finish_reason":"stop","message":{"content":"Hello there! How can I help you today?","role":"assistant"},"logprobs":null}]}%                         

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)

[membphis]

one minor issue

[membphis]

another style:

if access_token then
    return access_token
end

-- generate access token
... ...

[SkyeYoung]

I will check it when I have time.