Add abs() support for multivectors

[HarshRajSinghania]

Summary

Implement abs(mv) for kingdon multivectors.

MultiVector.__abs__ returns abs(self.norm().e), a non-negative scalar coefficient.

Motivation

Fixes #42. The issue requested that abs() work on multivectors. The maintainer clarified that the expected result is abs(mv.norm().e) as a scalar so comparisons such as points[abs(lines) < 1] work, and that this should follow norm rather than adding a new algebra operator.

Implementation

• Add MultiVector.__abs__ next to norm / normalized in kingdon/multivector.py.
• No new operator dictionary entry; abs() reuses norm() and then takes the absolute value of the scalar part (.e).

Testing

Added test_abs_multivector in tests/test_kingdon.py:

• Euclidean 2D vector (3, -4) → abs(v) == 5
• Negative scalar → positive result
• Comparison form from the issue (abs(v) < 6, not < 1)

Commands run locally (anywidget was unavailable in the test environment, so it was stubbed only to import the library):

pytest tests/test_kingdon.py::test_abs_multivector tests/test_kingdon.py::test_normalization

Result: 2 passed.

[tBuLi]

Hi @HarshRajSinghania, thanks for the PR!

The implementation looks good to me, but the tests seem to test several redundant things, and not the one thing that matters: can you add a test in e.g. Algebra(0,1) and test that the abs of a vector there is always positive?

And can I also ask you to change the masking example in arrays.rst to use this new syntax?

Thanks for your help.

[HarshRajSinghania]

Thanks @tBuLi — addressed both review notes:

• Added test_abs_algebra_01_vector_always_positive so Algebra(0, 1) vectors have abs(...) >= 0 for negative, zero, and positive coefficients.
• Updated the masking examples in docs/arrays.rst to use abs(l) / abs(x) instead of d.e comparisons, including the surrounding explanation.

Pushed to this branch: c8eaa882668ff422b3a4b5a38abcbcc78902cc8c.

[tBuLi]

I do not get the two assertions here, because we already know that result == 5.0. So these next two lines don't add anything and should be removed.

[HarshRajSinghania]

Removed the two redundant assertions after assert result == pytest.approx(5.0). The comparison cases and the Algebra(0, 1) positivity test are unchanged.

[HarshRajSinghania]

Removed the two redundant assertions in test_abs_multivector after result == 5.0, as requested.

[tBuLi]

Thank you @HarshRajSinghania for your effort, I have now merged the PR!