[AVX2, SIMD8_32u] Functions using AVX code

Some functions in SIMDVecAVX2_u<uint32_t, 8> have been copied from AVX for initial testing. Now since there are generic tests available, these should be re-factored to use AVX2 intrinsics. Functions are, among others, ADD & MUL. It would be good to fix them ASAP for performance.