Docs: Document dynamic model fetching feature in Add Model UI

[continue]

Summary

PR #12046 added a significant new feature that allows users to dynamically fetch provider models using their API keys.

Feature Description

• Added support for dynamically fetching provider models from various providers (OpenAI, Anthropic, Gemini, Ollama, OpenRouter, etc.)
• A refresh icon appears in the Add Model UI once an API key has been entered, allowing users to "update models list using my API key"
• Fetched models include proper context length, max tokens, and tool use capability detection
• Results are deduplicated with hardcoded models (hardcoded take precedence) and show proper provider icons
• Adds explicit support for Gemma 4 and GPT 5.4 family models

Documentation Needed

1. Document the new "refresh models" button in the Add Model flow
2. Explain that users can fetch available models dynamically using their API key
3. Document which providers support dynamic model fetching (OpenAI, Anthropic, Gemini, Ollama, OpenRouter)
4. Note that fetched model configurations (contextLength, maxTokens, capabilities) are automatically persisted to config.yaml

Related PR

#12046

Generated with Continue

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