AVX2 code adaption

---------------------
x86_64:
------------------------------------------
selfguided_3x3_8bpc_c: 308692.1
selfguided_3x3_8bpc_ssse3: 112436.5
selfguided_3x3_8bpc_avx2: 61749.8
------------------------------------------
selfguided_5x5_8bpc_c: 312132.8
selfguided_5x5_8bpc_ssse3: 79513.3
selfguided_5x5_8bpc_avx2: 45947.3
------------------------------------------
selfguided_mix_8bpc_c: 588951.9
selfguided_mix_8bpc_ssse3: 196751.5
selfguided_mix_8bpc_avx2: 109091.6
------------------------------------------
wiener_chroma_8bpc_c: 258874.8
wiener_chroma_8bpc_ssse3: 28172.4
wiener_chroma_8bpc_avx2: 16910.5
------------------------------------------
wiener_luma_8bpc_c: 264432.3
wiener_luma_8bpc_ssse3: 27958.3
wiener_luma_8bpc_avx2: 17303.8
------------------------------------------

---------------------
x86_32:
------------------------------------------
selfguided_3x3_8bpc_c: 350430.5
selfguided_3x3_8bpc_ssse3: 128850.8
------------------------------------------
selfguided_5x5_8bpc_c: 313963.6
selfguided_5x5_8bpc_ssse3: 81988.8
------------------------------------------
selfguided_mix_8bpc_c: 630584.2
selfguided_mix_8bpc_ssse3: 211802.0
------------------------------------------
wiener_chroma_8bpc_c: 288928.5
wiener_chroma_8bpc_ssse3: 30336.7
------------------------------------------
wiener_luma_8bpc_c: 284500.6
wiener_luma_8bpc_ssse3: 29521.9
------------------------------------------