FIX: Fix thread-safety for adaptive_minmax

[derb12]

Description

This PR fixes some issues with the testing framework when using threading, as found by using pytest-run-parallel. The majority of the identified test failures were just false-positives within the tests themselves since they were written without considering repeated, concurrent runs. The one true bug was that adaptive_minmax was found to not be thread-safe in both 1D and 2D, and pointed to a flaw with how the test_threading tests worked for optimizer-type algorithms, as detailed below.

Before this PR, the BaseTester class used the underlying private categorical algorithm classes (eg. _Whittaker, _Optimizer, etc.), rather than the Baseline class to test each method. This is fine for most algorithms, and the BaseTester.test_threading test added by PR #33 correctly identified most of the thread-unsafe behavior within the Baseline and Baseline2D classes. However, for optimizer-type methods, this will not find any issues with threading because when getting the baseline method to use, _Optimizers will always spawn a new object with the desired method (eg. _Polynomial if method is 'modpoly') since they do not have those methods, which will always be thread-safe. In actual use, the Baseline class will always have the desired method and will not spawn a new object. By changing the BaseTester[2D] test classes to use Baseline[2D] instead of the private classes, the test_threading test then correctly failed for adaptive_minmax in both the 1D and 2D cases, even without having to use pytest-run-parallel.

This oversight in the testing framework was fixed, as well as the underlying threading issues for adaptive_minmax for both 1D and 2D.

Note that linting is expected to fail, those fixes are already in the optimize_lam branch.

Type of Pull Request

• Bug Fix
• New Feature
• Miscellaneous Changes (refactor, code improvements, etc.)
• Documentation or Example Programs

Pull Request Checklist

• New code and/or documentation is valid for use with the BSD 3-clause license.
• New code is fully documented with docstrings that follow Numpy style,
  if applicable.
• New code follows PEP 8 standards as closely as possible, if applicable.
• Added/updated tests and ensured they pass locally, if applicable.
• Verified that documentation builds locally, if applicable.