support BASE_MODEL env var for 7B / larger model training

phase-2 fix: launcher was clobbering --reward_baseline when resume_from set

phase-2 tuning: anti-no-commit prompt + REINFORCE baseline env var

phase-2: add resume-from-checkpoint to GRPO training loop

phase-f fix: phase-1 domain coverage + checker robustness + flush + runs persistence

phase-f fix: drop pip install (UV venv has no pip; use PYTHONPATH=cwd)

phase-f: HF Jobs UV-script wrapper for T4 training