sensor raw16 data subtract 16 every pixel

We found that the every pixel data we get from v4l2 is 16 less than sensor output.

We want to know how xavier convert sensor raw16 data to memory.

We find the vi driver use “T_R16” to process raw16 data, but we cannot find how “T_R16” process raw16 data in TRM.

We guess the “16” is the black level, now we want to cancel this subtraction.

Duplicate with https://devtalk.nvidia.com/default/topic/1063832
Please trace with that.