Fix Issues Shane Discovered

[Grochocinski]

Fixed

High Minimum Ruby Version

The README claimed Ruby >=3.0 was required but we had no information on how to update Ruby included anywhere in the README. After taking a look at the datafarming gem, I believe we're good to drop the requirement down to Ruby >=2.5 (2.4 EOL'd 5 years ago, and 2.5 has been out for 7 years). Ruby can be installed a variety of ways and each has a unique way of being updated (plus it's different from system to system), so I'm hoping that dropping the requirement down to >=2.5 means we don't have to deal with that can of worms.

Unresponsive GUI

Issue: #124
I believe this was due to running all the experiment steps on the main thread, which prevented the GUI from handling any input in the meantime. I've introduced a fix for this that spawns a child thread to handle running each step of the experiment. This not only allows use of the GUI while an experiment is running, but it also allows for running multiple experiments at once. This is super useful if you have a short experiment you want to run but you already have a long experiment running up as you no longer have to wait for the long experiment fo finish.

Unsupported Tuple Operation Error

The new version of ASTRO-DF had an issue where two tuples were being subtracted from each other and that's not an operation supported by tuples. They are now copied into numpy arrays for that operation to occur.

Unclear Datafarming Errors

Shane reported a very generic error raised by the datafarming script despite having both Ruby and the datafarming module installed. I traced this error and believe it to have been caused by the terminal/IDE's path not updating after the datafarming gem was installed. I believe this error can be fixed by restarting the terminal/IDE, and I added that troubleshooting step into the error message.

TODO

MacOS Dark Mode

Issue: #125
MacOS's Dark Mode Overrides ttk theming. This makes the backgrounds black and makes the black text unreadable. I don't own to a Mac to fix this, but I should be able to gain access somewhere at NC State sometime today (12/12). If I can't figure out a solution, Shane said it worked fine once he switched over to light mode so we could just recommend participants do that.

Other Reported Bugs

Reported Hanging on GUI Launch

Shane that no window popped up for several minutes the first time he launched the GUI and that the hanging had something to do with matplotlib fonts. I wasn't able to find anywhere in the code that would use matplotlib other than the plots, which shouldn't be running on launch. I wasn't able to replicate this behavior but I'm very interested in seeing if conference participants experience something similar.

[Grochocinski]

I was able to replicate the slow first boot due to matplotlib building the font cache. It took <15 seconds and it only occurred on the first launch. We could probably add a warning about it potentially taking longer in the workshop writeup.