dungeon-escape/simple-v0: replace epsilon group-noise tier with per-group softmax-temperature-range tier (T~U(3.0,4.0) per cooperative group, step-reward norm 0.60)

feat(walker): simple tier, per-episode noise_std pool

feat(walker): medium tier, per-episode noise_std pool

feat(pushblock): simple tier, per-episode epsilon pool

feat(pushblock): medium tier, per-episode epsilon pool

feat(worm): simple tier, per-episode noise_std pool (continuous skill spread)

feat(worm): medium tier, per-episode noise_std pool (continuous skill spread)

feat(3dball-hard): simple tier, per-episode noise_std pool

feat(3dball-hard): medium tier, per-episode noise_std pool

add dungeon-escape/medium-v0 (Minari-aligned ~0.78, pool {0.42,0.44,0.46,0.48})

Add unity/pyramids/simple-v0 (stock Pyramids.onnx + epsilon=0.77 action noise; normalized return ~0.27)

Add unity/pyramids/medium-v0 (stock Pyramids.onnx + epsilon=0.70 action noise; normalized return ~0.51)

Add unity/dungeon-escape/simple-v0 (stock-r23 + per-group action noise; normalized skill ~0.27)

Add SoccerTwos simple-v0 1M dataset

Add unity/dungeon-escape/medium-v0 (stock-r23 + per-group action noise; normalized skill ~0.56)

rename soccer-twos/expert-selfplay-v0 -> soccer-twos/expert-v0 (drop algo suffix; expert-v0/medium-v0 convention)

rename dungeon-escape/expert-poca-v0 -> dungeon-escape/expert-v0 (drop algo suffix; expert-v0/medium-v0 convention)

Add SoccerTwos medium-v0 1M dataset

Add Walker release-23 expert trajectories

Add Worm release-23 expert trajectories

Add Pyramids release-23 expert trajectories

Redefine DungeonEscape and SoccerTwos as decentralized ego-agent datasets

Add release_23 SoccerTwos expert self-play dataset

Add discrete PushBlock expert Minari dataset

Add unity/3dball-hard/expert-v0 (stock ONNX policy, model-removed build)

Add unity/pushblock-with-input/expert-v0 (stock hybrid ONNX policy, model-removed build)

Add unity/crawler/medium-v0 (noise-synthesized tier from stock Crawler ONNX)

Add unity/crawler/simple-v0 (noise-synthesized tier from stock Crawler ONNX)

Add Minari trajectories unity/crawler/expert-v0 (1047 ep, 1M steps)

initial commit