Add SSSE3 implementation for pal_pred

pal_pred_w4_8bpc_c: 141.0
pal_pred_w4_8bpc_ssse3: 23.4
pal_pred_w8_8bpc_c: 374.5
pal_pred_w8_8bpc_ssse3: 29.0
pal_pred_w16_8bpc_c: 946.3
pal_pred_w16_8bpc_ssse3: 45.6
pal_pred_w32_8bpc_c: 1946.1
pal_pred_w32_8bpc_ssse3: 92.3
pal_pred_w64_8bpc_c: 4925.9
pal_pred_w64_8bpc_ssse3: 180.1