Switch GNMAnalysis to AnalysisBase

[PicoCentauri]

Fixes #3243

Changes made in this Pull Request:
Switch GNMAnalysis to AnalysisBase.

I don't think a versionchanged info is necessary here. For the user there are no breaking changes.

PR Checklist

• Tests?
• Docs?
• CHANGELOG updated?
• Issue raised/referenced?

[pep8speaks]

Hello @PicoCentauri! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

• In the file testsuite/MDAnalysisTests/analysis/test_gnm.py:

Line 140:80: E501 line too long (80 > 79 characters)

Comment last updated at 2021-04-28 11:41:50 UTC

[codecov]

Codecov Report

Merging #3244 (c09bec2) into develop (a23b1e9) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##           develop    #3244      +/-   ##
===========================================
+ Coverage    92.83%   92.84%   +0.01%     
===========================================
  Files          170      170              
  Lines        22809    22806       -3     
  Branches      3242     3241       -1     
===========================================
  Hits         21174    21174              
+ Misses        1587     1584       -3     
  Partials        48       48              

Impacted Files	Coverage Δ
package/MDAnalysis/analysis/gnm.py	99.18% <100.00%> (+2.38%)	⬆️

---

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 a23b1e9...c09bec2. Read the comment docs.

[orbeckst]

This looks like an uncontroversial, easy LGTM to me.

I looked at the 80% coverage for the diff and I don't think it's an issue. The only thing that's not tested is the case when the SVD fails and we skip a frame with an ugly print message.

Thanks @PicoCentauri .

@IAlibay would it be ok to squeeze it into 2.0? I'll add it to the milestone but leave it to you to remove it or to merge.

[IAlibay]

Couple of comments, I realise the print statement is a leftover of the previous code, but it would be much better if we could easily test it.

@orbeckst if one of the coredevs is happy to ensure that this gets merged on time for the 2.0 deadline then I have no objections. My comment on discord were more intented as a "please prioritize reviewing efforts on what is on the current to-do list" rather than a blanket code freeze.

[IAlibay]

Couple final comments (docs + matching on the warning text).

[IAlibay]

Docs do indeed build weirdly, looks like just a missing empty line though.

[IAlibay]

lgtm - @orbeckst are you happy with the recent changes?

[orbeckst]

Looks good to me. Thank you both! Please go ahead, Irfan

[orbeckst]

@IAlibay I'm jumping in and merging it — I think we're all happy with the state of the PR.