docs(readme): 📚 improve Antigravity provider feature documentation

Restructured the Antigravity provider description in the README for better clarity and readability:

- Converted the dense paragraph into a structured bullet list highlighting key features
- Separated thought signature caching, tool hallucination prevention, and thinking block sanitization into distinct points
- Replaced the informal troubleshooting note with a concise reference to dedicated documentation
- Added direct link to Antigravity documentation section for Claude extended thinking sanitization details

This change improves the discoverability of Antigravity's advanced features and provides a clearer path for users to understand Claude Sonnet 4.5 thinking mode limitations.

README.md

@@ -28,7 +28,11 @@ This project provides a powerful solution for developers building complex applic
 -   **OpenAI-Compatible Proxy**: Offers a familiar API interface with additional endpoints for model and provider discovery.
 -   **Advanced Model Filtering**: Supports both blacklists and whitelists to give you fine-grained control over which models are available through the proxy.
 
--   **🆕 Antigravity Provider**: Full support for Google's internal Antigravity API, providing access to Gemini 2.5, Gemini 3, and Claude Sonnet 4.5 models with advanced features like thought signature caching and tool hallucination prevention. However - Sonnet 4.5 Thinking with native tool calls is very skittish, so if you have compaction or switch the model (or toggle thinking) mid task - it will error 400 on you, as claude needs it's previous thinking block. With compaction - it will be destroyed. There is a system to maybe catch all this, but i am hurting my head here trying to come up with a solution that makes sense.
+-   **🆕 Antigravity Provider**: Full support for Google's internal Antigravity API, providing access to Gemini 2.5, Gemini 3, and Claude Sonnet 4.5 models with advanced features:
+    - Thought signature caching for multi-turn conversations
+    - Tool hallucination prevention via parameter signature injection
+    - Automatic thinking block sanitization for Claude models
+    - Note: Claude Sonnet 4.5 thinking mode requires careful conversation state management (see [Antigravity documentation](DOCUMENTATION.md#antigravity-claude-extended-thinking-sanitization) for details)
 -   **🆕 Credential Prioritization**: Automatic tier detection and priority-based credential selection ensures paid-tier credentials are used for premium models that require them.
 -   **🆕 Weighted Random Rotation**: Configurable credential rotation strategy - choose between deterministic (perfect balance) or weighted random (unpredictable, harder to fingerprint) selection.
 -   **🆕 Enhanced Gemini CLI**: Improved project discovery, paid vs free tier detection, and Gemini 3 support with thoughtSignature caching.