[ENH]: Add kdtw distance implementation

[SebastianSchmidl]

Reference Issues/PRs

What does this implement/fix? Explain your changes.

Adds KDTW distance (univariate, multivariate, and variable-length support).
Original implementation is at https://people.irisa.fr/Pierre-Francois.Marteau/REDK/KDTW/kdtw.m (Matlab ⚠️ ).

Open todos:

• current implementation is a similarity. How to convert to distance?
• generate expected values using original Octave implementation and fix tests
• replace kdtw kernel in KernelKMeans implementation

Does your contribution introduce a new dependency? If yes, which one?

no

[aeon-actions-bot]

Thank you for contributing to aeon

I would have added the following labels to this PR based on the changes made: [ $\color{#5209C9}{\textsf{distances}}$, $\color{#2C2F20}{\textsf{testing}}$ ], however some package labels are already present.

The Checks tab will show the status of our automated tests. You can click on individual test runs in the tab or "Details" in the panel below to see more information if there is a failure.

If our pre-commit code quality check fails, any trivial fixes will automatically be pushed to your PR unless it is a draft.

Don't hesitate to ask questions on the aeon Slack channel if you have any.

PR CI actions

These checkboxes will add labels to enable/disable CI functionality for this PR. This may not take effect immediately, and a new commit may be required to run the new configuration.

• Run pre-commit checks for all files
• Run mypy typecheck tests
• Run all pytest tests and configurations
• Run all notebook example tests
• Run numba-disabled codecov tests
• Stop automatic pre-commit fixes (always disabled for drafts)
• Disable numba cache loading
• Push an empty commit to re-run CI checks

[SebastianSchmidl]

This similarity-to-distance conversion part is problematic (if no input normalization is used) because the computed similarities are very, very small for most inputs and, thus, vanish.

E.g., all the following similarities will disappear in floating-point arithmetic (first two basic motions time series with different gammas):

• 1.0 - 2.7435e-123 == 1.0
• 1.0 - 4.9180e-72 == 1.0
• 1.0 - 6.5392e-53 == 1.0