Fix SandwichCompositor modifying input data

[djhoese]

In issue #1689 we realized that generating a true color composite while also generating a sandwich composite of that same true color produced an all black image. With this PR the SandwichCompositor no longer does an inplace operation.

One could argue that the utility methods used by the composites should always pass copies, but I think this is unnecessary in most cases and I'm scared it would hurt performance in the worst cases. This is the first time we've run into something like this...I think.

• Closes BackgroundCompositor using IR Sandwich (masked so only coldest clouds are visible) and True Color as inputs #1689
• Tests added

[codecov]

Codecov Report

Merging #1690 (01ae41a) into main (88938ab) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #1690   +/-   ##
=======================================
  Coverage   92.32%   92.32%           
=======================================
  Files         258      258           
  Lines       38123    38124    +1     
=======================================
+ Hits        35196    35197    +1     
  Misses       2927     2927           

Flag	Coverage Δ
behaviourtests	4.84% <0.00%> (-0.01%)	⬇️
unittests	92.86% <100.00%> (+<0.01%)	⬆️

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

Impacted Files	Coverage Δ
satpy/composites/__init__.py	88.74% <100.00%> (ø)
satpy/tests/test_composites.py	99.88% <100.00%> (+<0.01%)	⬆️

---

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 88938ab...01ae41a. Read the comment docs.

[coveralls]

Coverage increased (+0.0002%) to 92.812% when pulling 01ae41a on djhoese:bugfix-sandwich-inplace into 88938ab on pytroll:main.