Fix docker image building + CI improvements

[mikemhenry]

Developers certificate of origin

• I certify that this contribution is covered by the MIT License here and the Developer Certificate of Origin at https://developercertificate.org/.

Fixes #610

[codecov]

Codecov Report

Attention: 116 lines in your changes are missing coverage. Please review.

Comparison is base (7a39491) 92.21% compared to head (d011065) 90.97%.
Report is 8 commits behind head on main.

Files	Patch %	Lines
openfe/tests/protocols/test_openmm_afe_slow.py	25.39%	47 Missing ⚠️
openfe/analysis/plotting.py	12.82%	34 Missing ⚠️
openfe/protocols/openmm_rfe/equil_rfe_methods.py	48.64%	19 Missing ⚠️
...protocols/openmm_afe/equil_solvation_afe_method.py	60.00%	8 Missing ⚠️
openfecli/parameters/plan_network_options.py	93.75%	4 Missing ⚠️
openfecli/commands/gather.py	83.33%	3 Missing ⚠️
openfe/protocols/openmm_afe/base.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #641      +/-   ##
==========================================
- Coverage   92.21%   90.97%   -1.24%     
==========================================
  Files         124      127       +3     
  Lines        8594     8679      +85     
==========================================
- Hits         7925     7896      -29     
- Misses        669      783     +114     

Flag	Coverage Δ
fast-tests	90.97% <67.32%> (?)

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

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mikemhenry]

If I understand https://docs.codecov.com/docs/flags correctly, using flags should help with codcov comparing coverage from when we run slow tests (in the past) to PRs

[mikemhenry]

@IAlibay (and anyone else reviewing this)

This run tests the docker pipeline: https://github.com/OpenFreeEnergy/openfe/actions/runs/6897862437

So we want to make sure that passes before we merge this in.

[IAlibay]

overall lgtm, couple of comments

[IAlibay]

schedule is always a bit of a pain because codecov has that annoying thing where it fails if it gets the same commit too many times, would it make sense to limit this to merge instead?

[mikemhenry]

Ah, so its the merges! I couldn't figure out how (as the yaml existed) codcov got a bunch of the coverage for the slow tests since I noticed we didn't run the slow ones on a schedule, but we run the slow tests whenever it isn't a pull_request. Should I do not pr & not schedule to match the logic of when we run slow tests? Like if someone make a commit to main directly, I don't think that would be a merge event but the slow tests would still run.

[mikemhenry]

@IAlibay Let me know if the updated logic is better now, I went with "not schedule && not pull request" so we can cover any case where a slow test is ran (not pull) but only upload if it isn't a schedule run.

[IAlibay]

there's a syntax error on this line apparently

[mikemhenry]

can't weve comments into the expression it seems, fixed!

[IAlibay]

lgtm!