x86: Add smooth intra prediction AVX2 asm

Skylake-X benchmark results:

intra_pred_smooth_h_w4_8bpc_c: 149.9
intra_pred_smooth_h_w4_8bpc_avx2: 18.2
intra_pred_smooth_h_w8_8bpc_c: 425.9
intra_pred_smooth_h_w8_8bpc_avx2: 29.2
intra_pred_smooth_h_w16_8bpc_c: 1310.0
intra_pred_smooth_h_w16_8bpc_avx2: 64.4
intra_pred_smooth_h_w32_8bpc_c: 2144.9
intra_pred_smooth_h_w32_8bpc_avx2: 114.2
intra_pred_smooth_h_w64_8bpc_c: 4993.1
intra_pred_smooth_h_w64_8bpc_avx2: 231.6
intra_pred_smooth_v_w4_8bpc_c: 175.2
intra_pred_smooth_v_w4_8bpc_avx2: 17.6
intra_pred_smooth_v_w8_8bpc_c: 463.0
intra_pred_smooth_v_w8_8bpc_avx2: 25.2
intra_pred_smooth_v_w16_8bpc_c: 1353.9
intra_pred_smooth_v_w16_8bpc_avx2: 49.2
intra_pred_smooth_v_w32_8bpc_c: 2420.6
intra_pred_smooth_v_w32_8bpc_avx2: 107.2
intra_pred_smooth_v_w64_8bpc_c: 4649.8
intra_pred_smooth_v_w64_8bpc_avx2: 200.6
intra_pred_smooth_w4_8bpc_c: 307.6
intra_pred_smooth_w4_8bpc_avx2: 22.8
intra_pred_smooth_w8_8bpc_c: 631.0
intra_pred_smooth_w8_8bpc_avx2: 38.8
intra_pred_smooth_w16_8bpc_c: 1749.0
intra_pred_smooth_w16_8bpc_avx2: 97.6
intra_pred_smooth_w32_8bpc_c: 4493.1
intra_pred_smooth_w32_8bpc_avx2: 179.6
intra_pred_smooth_w64_8bpc_c: 8945.2
intra_pred_smooth_w64_8bpc_avx2: 370.7