[OptRed] Extend `-tritonintelgpu-optimize-reduction-locality` to support `repCluster[0] > 2`#2519

victor-eds wants to merge 1 commit intointel:fast-sub-group-transposeintel/intel-xpu-backend-for-triton:fast-sub-group-transposefrom

victor-eds:fast-sub-group-transpose-extendvictor-eds/intel-xpu-backend-for-triton:fast-sub-group-transpose-extendCopy head branch name to clipboard