Investigate dark model for 20260129, z0 for Matterhorn #1
Description
The z0 CCD was saturated doing pixflats in later Jan 2026. We flagged the bad data for 0128 where there was no science data. For 0129, we flagged the evening darks under the argument that they were now the brightest darks and were contaminating future night's dark models. However for the science data taken on 0129, those evening darks could help to better model the true dark current of the data taken on that night, since the current was presumably somewhere between what the evening darks measured and what the morning darks measured.
For daily we will live with what is on disk, which uses all darks including the bad ones on 0128. This is driven by no visible issues in the redshift QA for those bright and backup tiles. However, for Matterhorn, we should assess what to do for 0129 and whether to incorporate those evening darks or not. For now they will remain flagged as bad and not used.