Adds collapse_by_hour_of_day operator

[daflack]

Fixes #1029

Adds an operator to collapse by hour of the day to allow for diurnal cycle aggregation.

Contribution checklist

Aim to have all relevant checks ticked off before merging. See the developer's guide for more detail.

• Documentation has been updated to reflect change.
• New code has tests, and affected old tests have been updated.
• All tests and CI checks pass.
• Ensured the pull request title is descriptive.
• Conda lock files have been updated if dependencies have changed.
• Attributed any Generative AI, such as GitHub Copilot, used in this PR.
• Marked the PR as ready to review.

[github-actions]

[daflack]

Note for when reviewing: This works for pp files - where code is automatically read in to have 2 dim_coords that are associated with time: "forecast_period" and "forecast_reference_time" or if you wanted to split by hour for a single multi-day forecast, as in the tests.

However, in setting up the validity time equivalent branch and tests I noted that netcdf files are read in slightly differently to pp files. With netcdf files a CubeList with each cube representing a different forecast is created and that list could not be merged. I have since managed to work out how to recreate the type of cube required to allow aggregation over multiple case studies easily. I will apply this as a different issue and PR as this will influence the ability to aggregate over cases regardless of how it is done (e.g. hour of day, lead time, validity time).