Using querysets for add/remove_content methods

[daviddavis]

fixes #3814
https://pulp.plan.io/issues/3814

[codecov]

Codecov Report

Merging #3548 into master will increase coverage by 0.14%.
The diff coverage is 0%.

@@            Coverage Diff             @@
##           master    #3548      +/-   ##
==========================================
+ Coverage   57.44%   57.58%   +0.14%     
==========================================
  Files          59       59              
  Lines        2465     2459       -6     
==========================================
  Hits         1416     1416              
+ Misses       1049     1043       -6

Impacted Files	Coverage Δ
pulpcore/pulpcore/app/models/repository.py	55.07% <0%> (+0.78%)	⬆️
pulpcore/pulpcore/app/tasks/repository.py	37.03% <0%> (+4.77%)	⬆️

---

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 055c3da...1787314. Read the comment docs.

[bmbouter]

This all looks correct to me. I plan to rebase my code onto it for further testing post-merge. If you want formal pre-merge testing let me know, but I think Travis working is enough for me.

[dralley]

@bmbouter, @daviddavis I know this is something we don't really have bandwidth for, but it would be great to have a set of performance tracking tests to see the impact of changes like this, and future ones. I'm not sure how difficult that would be to do.

[bmbouter]

@dralley I'm a +1 on tracking that how our code changes affect the code performance over time. We have a performance tracking plan written up with some ideas. Overall I think we know what to track and how to do it, so we're blocked on time currently.