Upload folder using huggingface_hub

fix: batch normalization, EMA baseline, investigation bonuses, entity regex

fix: final audit — orchestrator scores, KL direction, grader typology alias

feat: self-play AML simulator — clean scenarios, reward composition, OS metrics

feat: Add PLR curriculum engine, GRPO trainer, and 5th AGUI panel; merge frontend updates

chore: Finalize Phase 1-3 upgrades (Memex UI, 3D Globe, Adversarial Training, Reward Caps)

feat: Phase 3 FinCEN 4-pillar expansion — industrial-grade AML environment

feature: initial AML implementation

resolved merge conflict