[ENH] Merge MultiRocket and MultiRocketMultivariate

[TonyBagnall]

merges the two classes MultiRocket and MultiRocketMultivariate. Part of #1699 fixes #1713

Unlike minirocket, I will keep the transform functions within multirocket separate and raise an issue about how the multivariate version is implemented. I think

1. it could be made more like the other rockets, but want to get the structure in place first.
2. there is huge duplication between mini and multi rocket, which can be rationalised at a later date
3. the random seed is set more often than needs be I think

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[aeon-actions-bot]

Thank you for contributing to aeon

I have added the following labels to this PR based on the title: [ $\color{#FEF1BE}{\textsf{enhancement}}$ ].

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 all pytest tests and configurations
• Run all notebook example tests
• Run numba-disabled codecov tests
• Stop automatic pre-commit fixes (always disabled for drafts)

[baraline]

Overall looking good, just got a question about methods (such as fit biases) that are still split between uni and multivariate, what is the reason exactly ?

Couldn't you just use the same method and affect channel_indices to channel 0 and/or num_channels_per_combination=1 for all kernels in the univariate case ?

[TonyBagnall]

Overall looking good, just got a question about methods (such as fit biases) that are still split between uni and multivariate, what is the reason exactly ?

Couldn't you just use the same method and affect channel_indices to channel 0 and/or num_channels_per_combination=1 for all kernels in the univariate case ?

you maybe could, but I want to do that on a separate PR for this one, I am not convinced as to how the multivariate is handled. See my comment at the top of the PR

[baraline]

Considering Tony's comment on rationalising common structures between rockets transformers later, I'd say that this PR has served its goal to put base structure in place.

[TonyBagnall]

I'll make a rocket related issue