Dataset + policies + eval for the **dose_solid** task (WetLabRoboData / ChemLab imitation-learning benchmark).