Fix Gemini Palm2 request config

[P3xill]

Summary

• Fixes the Gemini/PaLM-compatible client so Google generation settings are sent in generationConfig rather than Axios config fields.
• Exports Palm2AI plus typed Gemini request/response types from the AI package.
• Adds mocked Palm2/Gemini unit coverage, a testcases/palm2 Jsonnet prompt config, and a Jsonnet-driven palm2-chat example.

Bounty

/claim #279

Validation

• npx vitest run src/ai/src/tests/palm2/gemini.test.ts
• npm run build
• npx prettier --check JS/edgechains/arakoodev/src/ai/src/lib/gemini/gemini.ts JS/edgechains/arakoodev/src/ai/src/index.ts JS/edgechains/arakoodev/src/ai/src/tests/palm2/gemini.test.ts JS/edgechains/examples/palm2-chat/package.json JS/edgechains/examples/palm2-chat/src/index.ts JS/edgechains/examples/palm2-chat/tsconfig.json JS/edgechains/examples/palm2-chat/readme.md

Demo

The focused test demonstrates the request generated from a prompt and confirms that model, API key header, contents, response MIME type, token limit, temperature, topP, and topK are sent to the Google Generative Language API in the expected shape. The example keeps the prompt text in Jsonnet rather than hardcoding it in TypeScript.

[github-actions]

CLA Assistant Lite bot: Thank you for your submission, we really appreciate it. Before we can accept your contribution, we ask that you sign the Arakoo Contributor License Agreement. You can sign the CLA by adding a new comment to this pull request and pasting exactly the following text.

---

I have read the Arakoo CLA Document and I hereby sign the CLA

---

_{You can retrigger this bot by commenting recheck in this Pull Request}