[BUG] pyscope log keeps going on forever

[pgriffin17]

Describe the bug
The pyscope.log keeps going on forever without breaking into individual files for each day or session.

To Reproduce
Steps to reproduce the behavior:

1. Run telrun
2. Look at log

Expected behavior
The pyscope.log should be broken up into a new file occasionally, perhaps each day at noon (my preferred approach), or each session. The current one could still be called pyscope.log, and then get renamed after noon on that day if the first entry in the log file is in the previous day, or something like that.

Desktop (please complete the following information):
TCC1

[LLBlanc]

Note for people (maybe me too) who would want to work on this:

After much research, i.e. reading dreaded documentations, one of the best leads to solve this that I found was implementing the "TimedRotatingFileHandler" class from logging instead of the current "FileHandler" class.
https://docs.python.org/3/library/logging.handlers.html#timedrotatingfilehandler
However, it is not as straightforward as changing the logger config file to instantiate TimedRotatingFileHandler instead of FileHandler, as it requires a couple more arguments.
It would allow switching to a new pyscope.log file at regular intervals, and saving the old (e.g. previous day's) pyscope.log by renaming it with a timestamp. Additionally, it supports saving old logs indefinitely (don't delete the X latest pyscope.log) by setting backupCount to 0 (as in infinite amounts of backups), as well as having the switch happen based on UTC time, so new log switching and old log saving would happen at the same time for any and every environment running logging regardless of local time-zone.
However, the issue with this implementation arises with wanting to have this rollover happen at noon. The default rollover time is at midnight, which for the purposes of astronomy, would be highly inconvenient. However, to modify at what time the rollover happens, the setup needs to pass our desired time as a datetime object, not a date format string or something similar. This is a problem because we cannot explicitly pass a datetime object using config files.
Currently the only possible way I can see to continue down this path would be a little python/yaml magic, inserting the line "atTime: !!python/object/apply:datetime.time [12, 0, 0]" into the file logging handler config, which would correctly instantiate a datetime object. For this to work however, any instance of loading and parsing yaml configs would need to happen using the method unsafe_load instead of safe_load, which could introduce bugs, as well as cybersecurity vulnerabilities in a form similar to cross-site scripting (XSS).

[WWGolay]

This is a observatory operations specific issue with how we have our notebook that calls telrun operator set up, not an issue with pyscope. Closing (although still something we need to address)