[ENH] Add basic MORF implementation and design

[adam2392]

Towards #20

Need to also think about the design of the Cython API to allow future possibilities. Also porting over the oblique tree and related examples.

Before submitting

• I've read and followed all steps in the Making a pull request
  section of the CONTRIBUTING docs.
• I've updated or added any relevant docstrings following the syntax described in the
  Writing docstrings section of the CONTRIBUTING docs.
• If this PR fixes a bug, I've added a test that will fail without my fix.
• If this PR adds a new feature, I've added tests that sufficiently cover my new functionality.

After submitting

• All GitHub Actions jobs for my pull request have passed.

[adam2392]

@jshinm I will probably merge this in once I get the CIs working, but you may be interested in reviewing just the files in sktree/tree/tests/test_tree.py and docs/api.rst to i) get an idea of the breadth of unit-tests we want to see and ii) start brain-storming documentation.

The rest of the files are irrelevant unless you are interested in trying to improve the MORF trees.

[codecov]

Codecov Report

Base: 0.00% // Head: 83.66% // Increases project coverage by +83.66% 🎉

Coverage data is based on head (ebebdf4) compared to base (961b284).
Patch coverage: 83.80% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##            main      #21       +/-   ##
==========================================
+ Coverage   0.00%   83.66%   +83.66%     
==========================================
  Files          2        8        +6     
  Lines        101      557      +456     
==========================================
+ Hits           0      466      +466     
+ Misses       101       91       -10     

Impacted Files	Coverage Δ
sktree/ensemble/_supervised_forest.py	30.55% <30.55%> (ø)
sktree/ensemble/_unsupervised_forest.py	71.85% <71.85%> (ø)
sktree/__init__.py	77.27% <75.00%> (ø)
sktree/tree/_classes.py	88.78% <89.00%> (+88.78%)	⬆️
sktree/ensemble/__init__.py	100.00% <100.00%> (ø)
sktree/tests/test_forest.py	100.00% <100.00%> (ø)
sktree/tree/__init__.py	100.00% <100.00%> (ø)
sktree/tree/tests/test_tree.py	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[adam2392]

An issue seems that there is something up w/ Patch Oblique tree that fails on occasion:

=========================== short test summary info ============================
[406](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:407)
FAILED sktree/tree/tests/test_tree.py::test_sklearn_compatible_estimator[PatchObliqueDecisionTreeClassifier(random_state=12)-check_sample_weights_invariance(kind=ones)] - AssertionError: 
[407](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:408)
Not equal to tolerance rtol=1e-07, atol=1e-09
[408](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:409)
For PatchObliqueDecisionTreeClassifier sample_weight=None is not equivalent to sample_weight=ones
[409](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:410)
Mismatched elements: 1 / 16 (6.25%)
[410](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:411)
Max absolute difference: 1
[411](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:412)
Max relative difference: 1.
[412](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:413)
 x: array([1, 1, 1, 1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 2, 2])
[413](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:414)
 y: array([1, 1, 1, 1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 2, 1])
[414](https://github.com/neurodata/scikit-tree/actions/runs/4128531247/jobs/7133086102#step:14:415)
============= 1 failed, 262 passed, 34 skipped in 69.56s (0:01:09) =============

This could be due to resolving ties, or something, but it seems that it occurs only on 1 of the samples. Moreover, taking a look at https://github.com/scikit-learn/scikit-learn/blob/7970d78cfc1f8e705a8f417f29bb47b08e1bf0aa/sklearn/utils/estimator_checks.py#L1040, it seems that the last samples are essentially the same as the rest, so this could be a numerical issue from breaking ties or something.

Might be a deeper Cython issue where we need to figure out where some ties might occur.