Test Ukernel update

[Abhishek-Varma]

#define matmul_vectorized_c_func(lhs_ctype_in, lhs_mlir_type_in,                                             \
                                 rhs_ctype_in, rhs_mlir_type_in,                                             \
                                 acc_ctype_out, acc_mlir_type_out, M, K, N, r, s, t)                         \

@Abhishek-Varma Should the line above be M, K, N or M, N, K? The matmul_combos above this line seems to expect M, N, K instead?

Good find @erwei-xilinx ! I'll try with that!