[P] fastrad: GPU-native radiomics library — 25× faster than PyRadiomics, 100% IBSI-compliant, all 8 feature classes

PyRadiomics is the de facto standard for radiomic feature extraction, but it's CPU-only and takes ~3 seconds per scan. At scale, that's a bottleneck. I built fastrad — a PyTorch-native library that implements all 8 IBSI feature classes (first-order, shape 2D/3D, GLCM, GLRLM, GLSZM, GLDM, NGTDM) as n