Try to fix numpy dev0 install in nightly wheels

[IAlibay]

Fixes #

Changes made in this Pull Request:

PR Checklist

• [x] Tests?
• [x] Docs?
• [ ] CHANGELOG updated?
• [x] Issue raised/referenced?

Developers certificate of origin

• [x] I certify that this contribution is covered by the LGPLv2.1+ license as defined in our LICENSE and adheres to the Developer Certificate of Origin.

---

📚 Documentation preview 📚: https://mdanalysis--4383.org.readthedocs.build/en/4383/

[github-actions[bot]]

Linter Bot Results:

Hi @IAlibay! Thanks for making this PR. We linted your code and found the following:

There are currently no issues detected! 🎉

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (edda942) 93.41% compared to head (7af23d8) 93.41%. Report is 1 commits behind head on develop.

Additional details and impacted files

@@            Coverage Diff             @@
##           develop    #4383     +/-   ##
==========================================
  Coverage    93.41%   93.41%             
==========================================
  Files          171      185     +14     
  Lines        22511    23625   +1114     
  Branches      4129     4129             
==========================================
+ Hits         21028    22069   +1041     
- Misses         963     1036     +73     
  Partials       520      520             

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[IAlibay]

Let's see how CI looks, but I believe this should be good now. I swapped the order around in the hope that mamba would be more efficient at recognising the pre-installed (via pip) modules.

Actually on second thought, maybe the old approach was better so we could know if the dev0 build isn't picked up.

From the output of pip & mamba list it seems to be good.

[IAlibay]

Supplemented by https://github.com/MDAnalysis/mdanalysis/pull/4502