process_hd_data is awful

[klaragerlei]

I think this is correct, but I did it in a horribly confusing way and since get_hd_histogram is used in many places, this will be difficult to fix. The only reason it works is that the sampling rate of the ephys data = 1000 * the video sampling

https://github.com/MattNolanLab/in_vivo_ephys_openephys/blob/master/PostSorting/open_field_head_direction.py