Fix stride issue in c pack avx2 intrin

Created by: Cidana-Developers

In c_pack_avx2_intrin, the output stride is not consistent when the output data is stored. When the input width is 32, the output stride is fixed to 8; If the input width is 64, the output stride could be out_stride or 16. It would be good if we can use the out_stride consistently.