Remove emojis from notebook

Fix URL typos

Update train_colab.ipynb

Clean final training notebook: no demo, no Drive, all steps including plots + eval + download

Fix demo mode crash: use max_steps param instead of unpicklable local class

Optimize for fast iteration: 1.5B model, LoRA r=8, GRPO batch=2/gen=2, seq=512

Switch experiment tracking from W&B to TensorBoard (no API key required)

Enable W&B experiment tracking in SFT+GRPO phases (required by hackathon)

Data-Centric AI RL Environment — OpenEnv Hackathon Submission