figure out how to handle duplicate observations of the same galaxy

[rfinn]

The Problem

• the output directory for each galaxy is currently named cutouts/VFIDXXXX-NGCXXXX, or whatever the NED name is. We need to add something about the date of the observation so that observations from multiple dates don't overwrite each other.
• complicating factor is that the HDI coadds don't have date information passed through because I didn't set the header fields to propagate in swarp.

One Solution

• I edited Virgo/programs/build_web_cutouts.py
• user sets the telescope and run based on the directory that is getting processed
• downside is that the cutouts will stay stored as a subdirectory of the gui-output folder that notes the date, and they can't be merged into one cutout directory.
• a short-term advantage is that I don't have to rerun the gui analysis.

Alternate Solution

• change directory name created by the gui to include the telescope and run.
• advantage: is that this is would allow us to store all the cutouts in one directory without overwriting.
• disadvantage: is that I need to rerun the gui analysis for everything.
• this is the cleanest long-term solution.

[rfinn]

HDI coadds need to be created again anyway b/c I need to not have swarp do the sky subtraction. so this would favor the alternate solution. sequence of events:

• update halpha gui to include date and telescope in output directory name
  • this could be done in batch_gui if the information is not in the image headers
• recreate HDI coadds, using new naming convention, etc.
• move coadds to laptop
• solve for zp
• build psfs
• run through gui in auto
• run 2019 Feb datatthrough gui in auot