Add MCP•RL news section (#324)

Author: arcticfly

README.md

@@ -83,6 +83,7 @@ ART is an open-source RL framework that improves agent reliability by allowing L
 
 Explore our latest research and updates on building SOTA agents.
 
+- 🗞️ **[MCP•RL: Teach Your Model to Master Any MCP Server](https://x.com/corbtt/status/1953171838382817625)** - Automatically train models to effectively use MCP server tools through reinforcement learning.
 - 🗞️ **[AutoRL: Zero-Data Training for Any Task](https://x.com/mattshumer_/status/1950572449025650733)** - Train custom AI models without labeled data using automatic input generation and RULER evaluation.
 - 🗞️ **[RULER: Easy Mode for RL Rewards](https://openpipe.ai/blog/ruler-easy-mode-for-rl-rewards)** is now available for automatic reward generation in reinforcement learning.
 - 🗞️ **[ART·E: How We Built an Email Research Agent That Beats o3](https://openpipe.ai/blog/art-e-mail-agent)** demonstrates a Qwen 2.5 14B email agent outperforming OpenAI's o3.