Feature: Governance hooks for multi-agent role execution

MetaGPT assigns roles to agents that execute complex multi-step workflows. In production, teams need to track which role made which decision, enforce policies on tool access per role, and maintain an audit trail. asqav (pip install asqav) provides governance middleware that could wrap MetaGPT role actions with signed audit receipts and policy checks. This would make MetaGPT much more enterprise-ready without changing the core architecture. Would you be interested in a governance integration?