Mojo function

test_matmul_sm90

test_matmul_sm90[MType: CoordLike, NType: CoordLike, KType: CoordLike, //, a_type: DType, b_type: DType, c_type: DType, cluster_shape: IndexList[3], block_tile_shape: IndexList[3], wgmma_shape: IndexList[3], num_consumer: Int = 1, num_pipeline_stages: Int = 4, transpose_b: Bool = True, partitioned_multicast: Bool = False, grid_shape: OptionalReg[IndexList[2]] = None, use_tma_store: Bool = False, schedule: MatmulSchedule = MatmulSchedule.NONE, default_epilogue: Bool = False, elementwise_compute_lambda_fn: Optional[elementwise_compute_lambda_type] = None, measure_threshold: Optional[Float64] = None, backend: Backend = Backend.CUBLAS, k_group_size: Int = 1](ctx: DeviceContext, m: MType, n: NType, k: KType)