Mojo function

apply_epilogue_to_output_tile

apply_epilogue_to_output_tile[c_type: DType, c_tile_layout: Layout, //, elementwise_lambda_fn: fn[dtype: DType, width: Int, *, alignment: Int = 1](IndexList[2], mut SIMD[dtype, width]) capturing -> None, N: Int, WG_BN: Int, num_consumer_threads: Int, simd_size: Int](c_tile: LayoutTensor[c_type, c_tile_layout, MutableAnyOrigin, address_space=AddressSpace(3), alignment=128], c_gmem_wg_tile: LayoutTensor[c_type, layout, MutableAnyOrigin, address_space=address_space, element_layout=element_layout, layout_int_type=layout_int_type, linear_idx_type=linear_idx_type, masked=masked, alignment=alignment], c_gmem_wg_coord_m: Int, c_gmem_wg_coord_n: Int, local_thread_idx: UInt, M_bound: UInt32, N_bound: UInt32)

Apply the epilogue lambda function to the output data.

This function reads data from shared memory, applies the user-provided epilogue function (e.g., bias addition, activation), and handles bounds checking for edge tiles.