docs: Document dynamic provider model fetching feature

[continue]

Summary

PR #12046 added a significant new feature: dynamic model fetching for all providers. Users can now refresh and fetch available models from providers using their API keys.

Documentation Needed

Create documentation covering:

1. How to use dynamic model fetching: Explain the refresh button in the Add Model form that appears after entering an API key
2. Supported providers: List providers that support dynamic model fetching (Ollama, OpenRouter, Anthropic, Gemini, OpenAI-compatible)
3. What gets persisted: Document that contextLength, maxTokens, and capabilities (tool_use, image_input) are saved to config.yaml when adding a model
4. Auto-loading behavior: Explain that Ollama and OpenRouter models are auto-loaded on open

User-facing changes

• Refresh icon appears in Add Model form once an API key is entered
• Dynamically fetched models show in "Additional models" section
• Model capabilities are auto-detected and saved

Related PR

• feat: fetch provider models dynamically and fix isOpenSource pollution #12046 - feat: fetch provider models dynamically and fix isOpenSource pollution

---

Generated with Continue

Co-Authored-By: Continue noreply@continue.dev
Co-authored-by: bekahhw rhawrot@gmail.com