Jura: evaluate z7 late November 2022

In nightwatch, z7 was pretty bad on nights in late November 2022. In some cases these have been flagged as bad for the night (e.g. 20221120) while on other nights they were not flagged as bad. Perhaps the impacted fibers are flagged as bad at the fiber-level rather than fully masked at the CCD level, or perhaps the nightwatch masks were out of date compared to what is used for offline processing, or perhaps we have a real problem and need updated masks.

As part of Jura QA, investigate these cases more closely.

Example concerning image https://nightwatch.desi.lbl.gov/20221127/00155338/preproc-z7-00155338-4x.html:

Start with the Jura QA investigation of z7 nightwatch results for late November 2022, including the example night 20221127 and image 00155338. Compare nights such as 20221120 where the night was flagged with nights where it was not, and compare fiber-level flags and CCD-level masks with those used for offline processing. Done means explaining the discrepancy and determining whether updated masks or a real problem are involved.

Written by the indexing model from the issue text.