Circle ROIs, ROIs dimension selector and batch processing of videos

[Snapyou2]

I added the ability to use circles as ROIs, change the size in pixel of the ROIs, and also added a batch function to process multiple videos at once, with support for a different ROI shape for each video

[PolarBean]

looks good. Some small changes

• Could you please update the image in the readme to match your updates to the GUI.
• We should label both dimensions boxes explicitly, i see a label on just one and its not entirely clear what x in this context means.
• Not entirely neccessary but as i no longer have DLC installed it is hard for me to test this. I think for users as well it might be nice to include some example files that allow them to test the program.

Thank you for your update!

[Snapyou2]

Alright, thanks for feedback. I'll work on these improvements after my exam session, around 2 weeks from now.

[Snapyou2]

I have added your remarks into this PR, everything should be ok (had a lot of troubles with the README formatting haha, but it should be alright now).

[PolarBean]

great :) I see you've made more commits, let me know when it's ready for me to review and I will.

[PolarBean]

Would it be possible to include the mp4 you used for drawing the shape? Totally understandable if not!

[PolarBean]

Also wondering whether you would be interested in pursuing a submission to the Journal of Open Source software? I would be really happy for you to be first author and me last. It may also require a couple more features (Not as part of this pull request). Let me know if this is something you think would be worthwhile :)

[Snapyou2]

I could join the mp4 (maybe not the full one though, it is quite heavy), but I do not know what it would add really to the project except a few megabytes ? Maybe for demonstrative/tutorial purposes ?
And thanks for the journal submission idea but it was just a PR, I won't be able to invest much energy in this project these times. I think all my contributions are done, the last one is just a UI revamp because the workflow was not some clear and I updated the theming and sections of tkinter, it is mostly for aesthetic purposes, feel free to reject it if you do not like it.

[PolarBean]

I think the aesthetic changes are great. I think the mp4 would be nice for testing the roi drawing functions but its not necessary. I'll review this this week and merge it. Thanks for your efforts!

[PolarBean]

OK just tried to test without a video and hit an error. I think we need the video to test (a couple seconds is enough).

The steps I took were

1. create env
2. launch application
3. click Load DeepLabCut file.

This results in the error

❯ conda activate ROI_ENV
❯ python ROI_tool.py
Exception in Tkinter callback
Traceback (most recent call last):
  File "/home/harryc/anaconda3/envs/ROI_ENV/lib/python3.8/tkinter/__init__.py", line 1883, in __call__
    return self.func(*args)
  File "ROI_tool.py", line 130, in <lambda>
    self.LoadDeepLabfromFile = ttk.Button(dlc_analysis_frame, text="Load DeepLabCut File", command=lambda: self.posn_tracker.load_deeplab_Coords())
AttributeError: 'Application' object has no attribute 'posn_tracker'

This error may be due to having no video, however this isn't possible in the original DLC ROI Tool as it forces the user to open a video when the program launches. This will require one of two changes.

1. Force the user to open a video on launch.
2. Fix the above error so that users can open a DLC file without a video open.