[ENH] Update msm distance

[chrisholder]

Reference Issues/PRs

This pr depends on pr #367

What does this implement/fix? Explain your changes.

This pr updates msm to the new format for distances. In addition in doing so it makes optimisations for computing the distances between multiple time series.

Furthermore, this pr enables multivariate and unequal length time series to be used with msm. This is done by introducing the independent msm distance and dependent that the user can select which one they want to use with the independent parameter (when true will use the independent distance, when false will use dependent distance).

There was also a weird seg faulting bug for very large datasets with msm that would occur seemingly randomly and rarely. This also fixes this.

Finally there was a bug using the bounding window in msm that has also been fixed.

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

What should a reviewer concentrate their feedback on?

Did you add any tests for the change?

I've updated the existing tests for both independent and dependent msm and given them both univariate, multivariate and unequal length tests.

Any other comments?

PR checklist

For all contributions

• I've added myself to the list of contributors.
• Optionally, I've updated aeon's CODEOWNERS to receive notifications about future changes to these files.
• The PR title starts with either [ENH], [MNT], [DOC], or [BUG] indicating whether the PR topic is related to enhancement, maintenance, documentation, or bug.
• For new estimators, I've added the estimator to the online documentation.