separate boundary grad and value extrapolation, make grad default quadratic

[aabills]

Description

Issue #4433 pointed out that the boundary gradient is only first-order accurate. This is by default true, though the option to make it quadratic exists. I ran into this doing something else.

I tried setting it to be default quadratic, but this caused many test failures, often related to the diffusivity becoming negative when extrapolated. I'm guessing users have tuned parameter sets to just barely work with linear extrapolation:

In any case, linear extrapolation for boundary values is less of a priority than for boundary gradients.

So, this pull request separates extrapolation methods for boundary gradient and boundary value, and sets the default to be quadratic for the gradient. This solves the issue discussed in #4433 while not breaking other parameter sets.

Here is the modified graph from #4433

Fixes #4433

Type of change

Please add a line in the relevant section of CHANGELOG.md to document the change (include PR #) - note reverse order of PR #s. If necessary, also add to the list of breaking changes.

• New feature (non-breaking change which adds functionality)
• Optimization (back-end change that speeds up the code)
• Bug fix (non-breaking change which fixes an issue)

Key checklist:

• No style issues: $ pre-commit run (or $ nox -s pre-commit) (see CONTRIBUTING.md for how to set this up to run automatically when committing locally, in just two lines of code)
• All tests pass: $ python run-tests.py --all (or $ nox -s tests)
• The documentation builds: $ python run-tests.py --doctest (or $ nox -s doctests)

You can run integration tests, unit tests, and doctests together at once, using $ python run-tests.py --quick (or $ nox -s quick).

Further checks:

• Code is commented, particularly in hard-to-understand areas
• Tests added that prove fix is effective or that feature works

[kratman]

This should probably also be listed under breaking changes in the changelog, even though it probably will not impact many users

[aabills]

This should probably also be listed under breaking changes in the changelog, even though it probably will not impact many users

agreed, will do

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.21%. Comparing base (991a594) to head (8af9587).
Report is 2 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #4614   +/-   ##
========================================
  Coverage    99.21%   99.21%           
========================================
  Files          302      302           
  Lines        22857    22858    +1     
========================================
+ Hits         22678    22679    +1     
  Misses         179      179           

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

[brosaplanella]

Looks good, just a cosmetic suggestion in the CHANGELOG.

[kratman]

Re-approving after merging with the release branch