v0.0.9 - LLM Provider System Upgrade

What's New

🚀 LLM Provider System Upgrade

• 10 pre-configured providers: OpenAI, Google Gemini, DeepSeek, Groq, Together AI, Fireworks AI, OpenRouter, Mistral AI, Novita.ai
• Custom provider support: Add any OpenAI-compatible API with dynamic model fetching
• Model caching: 24-hour cache in GM storage for faster loading
• Smart baseURL detection: Auto-detect /v1, /v1beta, /openai paths

🛠️ Improvements

• Editable baseURL for all providers (with Reset button for presets)
• Add/delete custom providers from UI
• Auto-select first model when none selected
• Remove console.log in production builds

🐛 Bug Fixes

• Fix model selection not persisting
• Fix API endpoint construction for various providers