Add support for multiple values in the DecisionTree used for enhancements

[djhoese]

The current DecisionTree class, which is used as the base class for the EnhancementDecisionTree, currently has issues when a DataArray has multiple sensors specified for .attrs['sensor']. It currently causes a TypeError with a semi-cryptic debug log message produced:

Strange stuff happening in decision tree for %s: %s

This is because the DecisionTree expects a single string value for the "sensor" value, but is getting a set (ex. {'abi', 'glm'}). What this means is that you can never actually configure an enhancement for a DataArray with multiple sensors as it will fail when trying to find the match. This PR is meant to allow for you to both configure an enhancement for sensor: [abi, glm] to match it exactly, or for the provided DataArray with .attrs['sensor'] = {'abi', 'glm'} to match either sensor: abi or sensor: glm.

My one concern about this is that specifying sensor: [abi, glm] in the YAML config will make people think that that matches either abi or glm when it will actually only match the exact pair {'abi', 'glm'}.

• Tests added
• Passes flake8 satpy
• Fully documented

[codecov]

Codecov Report

Merging #1476 (1507caf) into master (744728f) will increase coverage by 0.20%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1476      +/-   ##
==========================================
+ Coverage   90.58%   90.78%   +0.20%     
==========================================
  Files         239      241       +2     
  Lines       34313    35111     +798     
==========================================
+ Hits        31081    31876     +795     
- Misses       3232     3235       +3     

Flag	Coverage Δ
behaviourtests	4.49% <38.73%> (-0.05%)	⬇️
unittests	91.25% <100.00%> (+0.19%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
satpy/tests/test_writers.py	98.88% <100.00%> (+0.33%)	⬆️
satpy/writers/__init__.py	87.30% <100.00%> (+0.79%)	⬆️
satpy/readers/seviri_l1b_hrit.py	91.90% <0.00%> (-0.28%)	⬇️
satpy/readers/_geos_area.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_abi_l1b.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_geos_area.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_vii_l1b_nc.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_seviri_base.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_fci_l1c_fdhsi.py	100.00% <0.00%> (ø)
satpy/tests/reader_tests/test_seviri_l1b_hrit.py	100.00% <0.00%> (ø)
... and 8 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 744728f...1507caf. Read the comment docs.

[mraspaud]

LGTM