feat(ai): integrate multiple AI providers and enhance configuration options

[narr07]

Summary

This PR introduces support for multiple AI providers (Google Gemini and Groq) alongside the existing Vercel AI Gateway in Nuxt Studio. Users can now choose their preferred AI model provider or allow Studio to auto-detect the provider based on available environment variables.

Key Changes

• Module Configuration: Enhanced ModuleOptions and runtimeConfig schemas to support provider-specific parameters:
  • Added provider option ('auto' | 'gateway' | 'gemini' | 'groq').
  • Added credentials and model configuration for Google Gemini (geminiApiKey, geminiModel, geminiFastModel).
  • Added credentials and model configuration for Groq (groqApiKey, groqModel, groqFastModel).
• Dynamic Provider Utility: Added src/module/src/runtime/server/utils/ai/provider.ts to dynamically resolve and initialize the appropriate AI client depending on active environment configurations.
• Server Routes Integration: Refactored /ai/analyze, /ai/commit, and /ai/generate server routes to leverage the unified provider utility instead of hardcoded AI Gateway client.
• Runtime Env Synchronization: Updated the studio-env middleware to resolve and load AI credentials from runtime environment variables (GEMINI_API_KEY, NUXT_STUDIO_AI_GEMINI_API_KEY, GROQ_API_KEY, etc.).
• Docs & Env Templates: Updated the AI documentation (docs/content/7.ai.md) and added placeholder options to .env.example files in the root and playgrounds.

How to Test

1. Set up a Gemini API Key (GEMINI_API_KEY) or a Groq API Key (GROQ_API_KEY) in your .env.
2. Configure ai.provider option inside nuxt.config (or leave it as auto to auto-detect).
3. Verify that the studio actions (AI-powered content analysis, generation, and commit messages) function correctly.

[vercel]

@narr07 is attempting to deploy a commit to the Nuxt Team on Vercel.

A member of the Team first needs to authorize it.

[pkg-pr-new]

npm i https://pkg.pr.new/nuxt-studio@538

commit: c739e8b