feat: add thesis section + Codex agent swarm narrative + 9B scaling probe + rewrite process log

feat: finalize swarm tooling and submission artifacts

docs: log QLoRA debugging and successful training start

feat: rewrite training to use TRL rollout_func + OpenEnv multi-turn pattern

docs: log Qwen 3.5 9B inference test on H100 (reward=0.25)

feat: fix openenv 0.2.1 API, add deployment files and GRPO training

docs: add process log and agent instruction