pyramids/medium-v0: re-record with the action-label gap fix (4.46% of rows held a placeholder 0); band U(2.7,4.9) unchanged

pyramids/medium-v0: fixed-T -> per-episode softmax-temperature-range tier (T~U(2.7,4.9) per episode, step-reward norm 0.8076)

pyramids/medium-v0: replace epsilon tier with softmax-temperature tier (T=3.7, step-reward calibrated)

feat(pyramids): re-collect medium with per-episode epsilon pool (trajectory diversity)

add pyramids/medium-v0 (Minari-aligned ~0.79, epsilon=0.59)

remove pyramids/medium-v0 (relabeled to simple; new 0.75 medium to be re-collected)

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