No magic cells in automatic run_tutorial.py and other related changes

[israelmcmc]

I realized, by eye and by chance, that one of the cells of the TS map tutorial was failing. Interestingly, the notebook continued running until the end, and the program reported a success.

While the error was only in a file checksum (fixed by this PR), and the code and tutorial are fine, this prompted me to look why this was not reported as an error.

As it happens, "cell magics" (the ones with "%%" at the beginning of the cell) are run in sub-process, and therefore the errors are caught inside and supressed. In particular, the TS map notebook is using the "%%capture" magic. Other notebooks use the "%%time" magic as well.

Prompted by this I changed a few things:

• Fixed the file checksum
• run_tutorials.py now looks for cell magics and comments them out.
• I added a test tutorial to check for this: test_must_fail_magic.ipynb
• Such tutorials with "must_fail" in the name now "fail succesfully". They show as FAILED but still in green.
• The program now catches the exceptions from failed tutorial, to allow the exceuted version to still be saved (up to the cell with the error)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.46%. Comparing base (79d2f92) to head (122d4ad).
⚠️ Report is 133 commits behind head on develop.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.