fix: use accelerate launch for DDP training instead of naive multi-GPU to avoid DataParallel crashes

fix: restore PyTorch compatibility by removing triton downgrade, rely on latest bitsandbytes

fix: downgrade triton to 3.1.0 in Kaggle notebook to fix bitsandbytes CUDA errors

fix: upgrade bitsandbytes to >=0.45.0 to resolve Triton ModuleNotFoundError on Kaggle

feat: add Kaggle notebook for Phase 3 training

initial deployment