v2.1: OpenEnv 0.2.1 integration + training notebook + HF Spaces deployment

v2.0: Novel world types + adversarial self-play for hackathon win

Hypothesis Engine: Scientific Discovery RL Environment for LLM Training