fix: eliminate all exact 0.0/1.0 paths, update reward_range in yaml and api to strict (0,1) bounds

Refactor for multi-mode deployment: move main.py to server/app.py

Fix env loading for inference and add reward_space to openenv.yaml

Update inference.py and openenv.yaml with improvements

Refactor GDPR Auditor: Full OpenEnv compliance & project cleanup