MOBT-799: CLI and plugin for subdividing duration diagnostics

[bayliffe]

Relates to: https://github.com/metoppv/mo-blue-team/issues/767

Adds a plugin to divide up duration diagnostics into shorter periods. Night or day masking can be applied to prevent spreading diurnally tied diagnostics into periods when they should not occur.

As an example, splitting 3-hours of sunshine duration, held in a 6-hour period cube, into two 3-hour period cubes. If the day/night terminator for a grid point being considered fell across the site at exactly the mid-point of the original 6-hour period, all 3-hours of sunshine duration will be allocated to the 3-hour period that falls in day time. If this were the mid point of the day the duration would simply be divided up equally into the two shorter periods.

This code cannot take account of weather impacts, e.g. cloud on sunshine duration. It is intended for use in constructing daily summaries (24-hour periods), but making these available at 3-hourly intervals to serve global time zones. The method will introduce errors, and limits the accuracy of the data to the accuracy of our simple 2-D day/night mask. These are compromises made to serve global time zones when the input data is only available at 6-hourly intervals from ECMWF.

Testing:

• Ran tests and they passed OK
• Added new tests for the new feature(s)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.41%. Comparing base (84a8944) to head (6c1e486).
Report is 58 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2060      +/-   ##
==========================================
+ Coverage   98.39%   98.41%   +0.01%     
==========================================
  Files         124      134      +10     
  Lines       12212    13227    +1015     
==========================================
+ Hits        12016    13017    +1001     
- Misses        196      210      +14     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[gavinevans]

Thanks @bayliffe 👍

I've added some minor comments below.

[gavinevans]

Thanks for the updates, @bayliffe 👍

I've added a comment related to one typo.

[Kat-90]

I've read through and understood the code and tests, the comments in the unit tests were really helpful.
The tests on my local branch all pass.
I have requested some changes to the doc strings based on my understanding of it and one suggestion for another test.

[bayliffe]

@Kat-90 I've responded to the review. Please take a look at the acceptance test data too.

[Kat-90]

Thanks for the changes. I've run the new tests, LGTM.