Conditionally determine visibility of columns in datatables

[HarmvZ]

This PR allows us to conditionally show / hide columns in a datatable.

It works by defining an optional_condition callable on a Column definition for datatables. The optional_condition will be called with the object for that row as parameter for each visible row on the page. If non of the calls evaluate to True, the column will be hidden. The visibility is recalculated each time the table data changes: pagination changes, search changes, ordering changes, etc.

This will be useful when the fields in DIAGNijmegen/rse-panimg#51 are implemented here and we want to show these columns in the archive cases list view, but hide the columns that contain empty values.

As an example I made the "Algorithm Results" column in the cases list view only visible if it actually contains algorithm results.

[jmsmkn]

At a quick glance, I think the way this is implemented has an N+1 problem with 2 extra calls per object. Have you checked by using the debug toolbar, including the history tab to monitor the AJAX calls?

[HarmvZ]

By the optional_condition for algorithm results? The data for that is prefetched in get_queryset so it should not need extra calls to get it. The toolbar indicates 14 queries total independent of the page length / nr of objects.

[codecov]

Codecov Report

Merging #2044 (6dc7463) into master (51b99c8) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #2044   +/-   ##
=======================================
  Coverage   93.42%   93.43%           
=======================================
  Files         692      692           
  Lines       26452    26456    +4     
=======================================
+ Hits        24713    24718    +5     
+ Misses       1739     1738    -1     

Impacted Files	Coverage Δ
app/grandchallenge/archives/views.py	88.67% <ø> (ø)
app/tests/core_tests/test_views.py	100.00% <ø> (ø)
app/grandchallenge/datatables/views.py	100.00% <100.00%> (ø)
app/grandchallenge/components/models.py	93.08% <0.00%> (+0.28%)	⬆️

---

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 51b99c8...6dc7463. Read the comment docs.

[jmsmkn]

Great! Then my mistake :-)