Fix xdl bug

Author: EnricoDeg

include/ck/tensor_operation/gpu/grid/gridwise_gemm_bias_add_reduce_xdl_cshuffle_v1.hpp

@@ -897,6 +897,8 @@ struct GridwiseGemmBiasAddReduce_k0mk1_k0nk1_mn_xdl_cshuffle_v1
             static_assert(num_access == sfc_c_global.GetNumOfAccess(), "wrong!");
 
             static_for<0, num_access, 1>{}([&](auto access_id) {
+                block_sync_lds();
+
                 // each thread write its data from VGPR to LDS
                 c_thread_copy_vgpr_to_lds.Run(c_thread_desc_m0_n0_m1_n1_m2_m3_m4_n2,
                                               sfc_c_vgpr.GetIndexTupleOfNumber(access_id),