d207/d63/d45/d117/d135/d153

~9-45% better depending on the predictor on 32-bit ARM, similar range on
x86-64

this matches the non-highbitdepth implementation

BUG=webm:1316

Change-Id: Iddebdf7c58c6f31c47cae04da95c6e5318200e4c