feat: implement Unsloth GRPO training script with diverse reward functions and logging

Merge branch 'main' of https://github.com/LO-Kyu/gridmind

remove binary from tracking

feat: enable huggingface-skills plugin in cursor settings

fix: update training script with seed variation, fix reward normalization, regenerate training curves showing 0.52->0.67 improvement

feat: add scripts/full_demo.py — unified 10-step demo proving all 4 hackathon themes operational

docs: add HF blog post draft for community posting

fix: training reward uses 8-step rollout + /grade for genuine episode-level signal

feat: commit training evidence, update README with real scores, add demo scripts

feat: add baseline evaluation tools and demo scripts for RL performance comparison

chore: ignore ML outputs, models, and AI agent artifacts

fix: include env/faults.go missed due to .gitignore

fix: add missing faults.go file causing HF space build error

feat: add GridMind GRPO training notebook using Unsloth and HF TRL

feat: implement Go-based GridMind-RL simulation core and update inference interface (graph)

docs: add reward weight rationale table to README

feat: add multi-agent and planning CLI flags to inference and expose environment metadata via /info endpoint

feat: define GridMind-RL environment data models and task structures

feat: implement multi-component dense reward function and environmental logic for GridMind-RL

Add Task 4 instruction following, Curriculum Manager for self-improvement, and world modeling simulation

Add GridMind GRPO training notebook for Colab

fix: Replace ineffective break with return in WebSocket close handler

feat: Add full OpenEnv compliance

refactor: Inference and readme

updated readme

rewrite reward system

fix llm based reward output

fix: problems with parameters

fix: reward system

fix: introduce SCORE_EPSILON and clamp scores in run_episode and main functions

fix: clamp scores after rounding and ensure all sub-scores are clamped

fix: clamp all scores to open interval (0, 1) to meet validator requirements

fix: mirror all base images via gcr.io

fix: use gcr.io mirror to avoid Docker Hub rate limits

fix: use golang:1.21 instead of alpine for better Docker registry compatibility

fix: auto-start env server in inference.py

refactor: update default model and API endpoint, enhance error handling, and add close method for compatibility

Merge branch 'origin/main' with local changes overriding

fix: provide fallback API key and add safety check for empty observations in inference client

Sync root inference.py with fixed python/inference.py

Fix inference.py: handle missing API key gracefully, wrap all exceptions

fix: add 'server' entry point for OpenEnv validator

fix: add server entry point and pyproject scripts for OpenEnv validator

Merge branch 'main' of https://github.com/LO-Kyu/gridmind

Fix UTF-8 encoding for landing page emojis

Fix GitHub URL in landing page

Fix supervisord permissions for HF Spaces non-root user

Add root landing page handler with links to dashboard and API endpoints

Fix dashboard paths for /dashboard reverse proxy

Add dashboard reverse proxy at /dashboard