Document feature extraction graceful degradation

[wbwakeman]

As a data manager for the next release, I want to understand the reasons why any individual candidate cell may or may not be suitable for release, so that I can make a decision to release it or not.

To that end, I want to understand the changes that were introduced in #428 (and related issues #398, #278, #324, #330, #394) so that I can clearly understand the conditions that will cause a cell to pass or fail feature extraction, whether or not a sweep passes or fails QCs, and what level those criteria apply (e.g. cell, sweep, etc.)

Cells can fail QC but still run feature extraction
Sweeps can be removed, and are then ignored
Feature extraction can fail on sweep categories.

Tasks:

• [ ]

Acceptance criteria:

• Provide link to clearly written documentation