Sync updates from upstream

[utensil]

Addresses #505 .

This PR aims to:

• incorporate the files from upstream, as is when possible, but avoid breaking compatibilities within this PR
• lint the code
• add the notebooks from upstream and use them as tests
• markers like GSG code starts/ends will be kept to make it easier to identify the added code and distinguish it with other changes between 0.5.0 and 0.5.1, as well as GSG's other proposed changes
• raise discussions about the changes the break previous tests
• add documentations for new APIs (there are basic docs, I choose to use follow-up PRs to improve)

The goal here is to make it a foundation for merging other people's work, and avoid accidentally rolling back previous work (check v0.5.0...v0.5.1 to decide what to keep when there is code conflict).

The strategy is to add them and fix them file by file.

TODOs:

• Add Python files
  • Add mv
  • Add lt
  • Add gprinter
  • Add GAlgebraInit (renamed to primer)
• Add notebooks (They are put into examples/primer, and GAlgebraOutput.ipynb is renamed to gprint.ipynb, and the rest are ensured to be in lowercase.)
• Ensure test coverage (with primer notebooks: -1.65%; with undual tests: -1.10%; with lt tests: -0.55%; with mv tests: +0.38%)
• Add change log
• Check new API doc
• Investigate the diff of lin_tran_check
• Investigate the diff of test_gsg_undual_etc.ipynb
• Add the migration guide for the readers of LAGA&VAGC

[utensil]

The new norm break the following test:

    def test_abs(self):
        ga, e_1, e_2, e_3 = Ga.build('e*1|2|3', g=[1, 1, 1])
        B = ga.mv('B', 'bivector')
        assert abs(B*B) == -(B*B).scalar()

Seems fixable by using metric.square_root_of_expr instead of sqrt

[codecov]

Codecov Report

Attention: Patch coverage is 75.91623% with 92 lines in your changes are missing coverage. Please review.

Project coverage is 78.95%. Comparing base (d4d09bd) to head (0defe0c).

Files	Patch %	Lines
galgebra/lt.py	75.33%	37 Missing ⚠️
galgebra/gprinter.py	74.16%	31 Missing ⚠️
galgebra/primer.py	0.00%	15 Missing ⚠️
galgebra/mv.py	89.74%	8 Missing ⚠️
galgebra/metric.py	94.11%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #510      +/-   ##
==========================================
+ Coverage   78.05%   78.95%   +0.90%     
==========================================
  Files          17       19       +2     
  Lines        4196     4400     +204     
==========================================
+ Hits         3275     3474     +199     
- Misses        921      926       +5     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[utensil]

CI is green for mv.py except dropped test coverage. d853c70

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[utensil]

The changes to test outputs of simple_ga_test are just using covariant/contravariant notation instead of writing all indexes as subscripts, this was suggested in #475 .

The changes to test outputs of lin_tran_check is much significant and looks suspicious:

Will investigate later. The current green CI is a delusion for now. The visual diff is at: https://app.reviewnb.com/pygae/galgebra/commit/33a33fbefc67a480e7f0565273cc16b0e34f8d09/