feat: add SFT loss comparison visualization and training notebook for Qwen models

refactor: update JSONL files for e1 launch readiness and h1 acquisition defense scenarios

refactor: update training scripts and documentation for SFT and RLVR processes

feat: add environment setup and requirements for Lightning AI H100 training

feat: update README and runbook for SFT and GRPO training enhancements

feat: add summary generation and visualization for model evaluation results

feat: update evaluation results and training scripts for Qwen2.5-7B-Instruct

comm

feat: add new task definitions and data files for launch readiness scenarios

feat: add evaluation and plotting scripts for CORP-ENV