Adds script to run notebooks and catch any errors

[aamster]

#2601

• Adds script to run notebooks and catch any errors
• This script is run as part of a new github actions workflow which is triggered when the target branch is "master". Example run: https://github.com/AllenInstitute/AllenSDK/actions/runs/3498866966/jobs/5859760665. Many notebooks are failing. Some won't run until we update data on S3. Created Fix issues in notebooks #2611 to fix the others.
• The notebooks parameterize the output_dir to manage files output by the notebook using a tag named "parameters". This is in the papermill (library used to run notebooks) docs. Before, these paths referenced internal paths or fake paths, and so the notebook wouldn't run.
• Comments out blocks of notebook that output excessive data (~1 TB) or take an excessive amount of time to run (~1 hr)
• Skips behavior_ophys_session notebook since it can only be run internally. Didn't want to add a 2.5 GB NWB file to the repo or change the notebook to use the cloud api.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[mikejhuang]

What version of python would this use?

[aamster]

The default version of python that ships with the "ubuntu-latest" image, which currently is 3.8, but looks like it will be 3.10 on december 1:
https://github.com/actions/runner-images/blob/main/images/linux/Ubuntu2004-Readme.md

[mikejhuang]

It was interesting to see how notebooks can be parameterized and automated with papermill.

[aamster]

@mikejhuang thanks for feedback. I think I addressed all your feedback.

[mikejhuang]

Looks good!
What if the output of the notebook changes without throwing errors? Is there a way to review it?

[aamster]

What if the output of the notebook changes without throwing errors? Is there a way to review it?

@mikejhuang the notebook changes are automatically committed, like in this commit . But we can review the auto commit.