Problems with the `reverse` attribute

[klapo]

The reverse attribute was used in too many places with different meanings, leading to some of the problems with section labels that decrease in LAF.

There needs to be a robust way to handle the reverse flag used in calibration separately from the label indicating if a sections LAF to physical coordinate mapping is reversed vs the reverse flag for an entire channel when doing double-ended calibration.

My idea is to have some sort of attribute for each section that is a dictionary indicating the relative direction of the coordinates in LAF space. This would then free up the reverse attribute to be consistent with the usage in the dtscalibration library.