[BUG] SFA/SFB s2t (UTCCP) should be issued only by the leader CTA

Which component has the problem?

CUTLASS C++

Bug Report

Describe the bug
At cutlass/gemm/collective/sm100_blockscaled_mma_mixed_tma_cpasync_warpspecialized.hpp:L1029-1032, s2t copy is performed by both SMs when 2SM mode is enabled.

Steps/Code to reproduce bug

This itself seems to work correctly, but has a potential race condition bug.

Expected behavior
The stub should be protected by a branch that checks is_mma_lead_cta i.e.

    if (is_mma_leader_cta && cute::elect_one_sync()) {
      copy(tiled_copy_s2t_SFA, thr_tCsSFA_s2t(_,_,_,_,read_stage_tma), thr_tCtSFA_s2t);
      copy(tiled_copy_s2t_SFB, thr_tCsSFB_s2t(_,_,_,_,read_stage_tma), thr_tCtSFB_s2t);
    }

Environment details (please complete the following information):

• Environment location: B200

Start at cutlass/gemm/collective/sm100_blockscaled_mma_mixed_tma_cpasync_warpspecialized.hpp:1029-1032 and trace the 2SM path around the s2t SFA/SFB copies. Check how is_mma_leader_cta and elect_one_sync are used nearby, then validate the B200 reproduction or relevant 2SM tests. Done means the copies are issued only by the leader CTA without introducing a race.