…ck/warp shape size, this now touches the perf of fusedL2NN simt kernel

… from per row to multi-rows. now this kernel is 1.3x to 1.8x faster as k value increases perf gets better than fusedL2NN simt kernel

… to predicated tile iterator

… than per warp multi row lock, cleanup and doc update

… tile iterator, this both improves perf by 20%+ compared to previous version by reducing gmem atomics+coalesced stores

…ing raft_cuda_try

…election, add comments on register spills tile shape, add test case for veclen=2

…usedL2NN_cutlass

…fix cutlass_utils.h comments

…ing 2 copies of the same code

…ntime

…usedL2NN_cutlass

…usedL2NN_cutlass

… accordingly