V4: steep reward slope - full-episode eval + stronger milestones + progress bonus

fixes : rewards and training

Fix reward consistency and timeout progress

Fix reward-stationary reset controls and notebook training setup

fix: rebalance reward range + unblock enrollment->monitoring->analysis phases

fix: remove duplicated curriculum block, dedup FDA table, accumulate total_reward for invalid steps

fix(N1-N6): unblock SUBMIT_TO_FDA, harden episode lifecycle, add tests

V4: fix 5 critical + 8 major + 2 moderate issues

fix: V2 reward tuning + single-step training script for GRPO slope

fix: rebalance rewards for GRPO slope - remove efficiency baseline, add milestone bonuses

Merge branch 'Roopalgn:main' into main

fix: G26 wire trainer.train(), G23 fix episode_phase progression

fix: populate latent biology and failure telemetry

fix: add root route so HF Space doesn't show 404

Push 7 Phase A: grounding + kaggle notebook + docs

style: fix ruff lint and format errors

fix: wire shaping_bonus, advance_curriculum, AdversarialDesigner into episode_manager

[Push 6] Suyash: CI workflow, container hardening, /transcripts endpoint, entrypoint validation, lint/format fixes

[Push 5] Suyash: adaptive curriculum, dashboard backend, hardening

[Push 4] Suyash: train.py, eval_compare.py, plot_rewards.py, env config, LLM judge

Push 3: Curriculum controller, hidden-state pipeline, phase detector, trial judge, and full EpisodeManager wiring

Push 2: Rule-Engine, Reward Components, Episode Logging, Noise Model

fix(structure): align file layout with ARCHITECTURE.md — rules/ and root models.py

refactor(env): rename environment/ to server/, add openenv.yaml, fix Dockerfile base image