StainX: GPU-accelerated batch stain normalization for computational pathology at scale

StainX implements histogram matching, Macenko, and Reinhard normalizers via a portable PyTorch backend and an optional CUDA backend that fuses per-pixel operations for batch throughput. Built around a two-stage fit/transform interface, it addresses the lack of fused multi-image batch transforms in existing cohort-scale pipelines. Performance on NVIDIA GPUs suggests significant speedups for large histopathology datasets. This tool could standardize stain normalization across computational pathology workflows, facilitating more reproducible large-scale analyses.