0.3.1

[chandralegend]

• Added the State Machine module to manage complex workflows and state transitions.
• Interactive Training Module for agents, allowing users to train agents interactively with feedback.
• Improved Memory module with Decision Model Caching for better performance and reduced latency.
• Ability to add type definition to functions using the config.
• Anthropic LLM provider support.
• Better Type Validation for Decision Models.
• Better Error Handling for Decision Models by Constraining the Fields according to the errors.

---

This discussion was created from the release 0.3.1.