fix: enforce strict (0, 1) range for task scores and rewards

chore: compliance fixes for OpenEnv RL Challenge

Elevate Creativity & Real-world utility: Implement interactive multi-turn Simulator & Tool usage mechanics

Update inference loop to correctly output Hackathon validation stdout formats

Fix environment score accumulation scaling bug which artificially inflated bounds

Fix baseline score accumulation and restrict openenv metadata range constraints

Fix phase 2 validation error strictly bound scores and add has_grader to tasks

Bound cumulative and step rewards strictly between 0 and 1

Fix Phase 2 grader bounds strictly between 0 and 1

fix: resolve Phase 2 failure by adding requests dependency and improving inference script robustness

fix: rebuild inference.py with correct logging and env vars

fix: clean up requirements.txt formatting

fix: add requests to root requirements.txt

fix: add requests dependency

fix: inference.py sends structured Action JSON to /step endpoint

Add uv.lock for multi-mode deployment

fix: add HEAD handlers for / and /health so HF Space turns green

chore: trigger HF Space rebuild

fix: add HTML root route so HF Space shows Running (green) status

Cleanup: untrack .DS_Store and uv.lock files

Fix: remove malformed colon-path entries causing HF Space failure

Fix /reset to handle null body correctly

Force rebuild to bust HF cache

Add openenv validate required files

Stop tracking __pycache__ files

Fix POST /reset validation for missing JSON bodies

Fix inference.py RTF formatting error

Fix /reset to accept empty body using Body(default=None)

Fix openenv.yaml author placeholder

Fix /reset endpoint 422 error - accept JSON body

Update app:main.py

Fix HF Space README configuration

Add GitHub Actions deploy workflow

Create deploy.yml

Revise team information in README

Final Polish: Tasks, Graders, and Root Inference script

Emergency Fix: Populated all task logic files

Initial commit: Production-ready OpenEnv backend