Wet bulb freezing level

[mspelman07]

This PR adds in the ability to calculate a Wet bulb freezing level from a wet-bulb temperature on height levels.

Most of the changes for this PR are simply renaming variables to be more agnostic and apply to height and pressure coordinates rather than just pressure. The main changes are for the plugin ExtractPressureLevel (renamed to ExtractLevel). The core changes in this PR are:

• Plugin now looks first for a pressure coordinate and then a height coordinate. Depending on which it finds will determine whether a height or pressure level are extracted.
• In writing this plugin I found that stratify has an EXTRAPOLATE_NEAREST keyword which will automatically do what fill_in_bounds was previously doing so I have replaced this.
• I have also added an extra argument to the stratify interpolation of rising=False. This tells stratify which direction we expect the variable to change and solves a problem if there was a temperature inversion between the first two valid points (see notebook below)
• CLI added to calculate wet bulb freezing level

To show my changes to the ExtractLevel plugin haven't changed how the plugin works (except where they should change it) I have plotted a few profiles of what it extracted previously and what it extracts now (https://github.com/MetOffice/improver_aux/blob/extractpressure_behaviour/notebooks/Checking_changes_to_ExtractPressureLevels.ipynb)

Acceptance_test_data: metoppv/improver_test_data#30

Testing:

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

[codecov]

Codecov Report

All modified lines are covered by tests ✅

Comparison is base (64b8eab) 98.38% compared to head (120518c) 98.38%.
Report is 2 commits behind head on master.

❗ Current head 120518c differs from pull request most recent head 96e8638. Consider uploading reports for the commit 96e8638 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1949      +/-   ##
==========================================
- Coverage   98.38%   98.38%   -0.01%     
==========================================
  Files         123      123              
  Lines       11792    11783       -9     
==========================================
- Hits        11602    11593       -9     
  Misses        190      190              

Files	Coverage Δ
improver/psychrometric_calculations/hail_size.py	100.00% <100.00%> (ø)
improver/utilities/cube_extraction.py	100.00% <100.00%> (ø)

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

[MoseleyS]

Just a doc-string change needed.

A comment for your notebook: Surface temperature inversions like your final example are quite common, although they are usually much shallower than that. We won't have seen this in the hail output because surface temperature inversions are associated with stable, nocturnal, non-convective weather, when hail is not possible.

[MoseleyS]

Suggested change

	The height level at which the wet-bulb temperature first drops below 237.15K
	The height level at which the wet-bulb temperature first drops below 273.15K

[mspelman07]

Done

[MoseleyS]

Suggested change

	In grid squares where the temperature never goes below 237.15K the highest
	height level on the cube is returned. In grid square where the temperature
	starts below 237.15K the lowest height on the cube is returned.
	In grid squares where the temperature never goes below 273.15K the highest
	height level on the cube is returned. In grid square where the temperature
	starts below 273.15K the lowest height on the cube is returned.

[mspelman07]

Done

[brhooper]

Thanks @mspelman07, I've just added a few comments, mostly about docstrings.

[brhooper]

Thanks @mspelman07, I'm happy to approve this now