testing: fixtures: fix docker regression

[efiop]

Regression from #9054

[codecov]

Codecov Report

Base: 92.97% // Head: 92.66% // Decreases project coverage by -0.31% ⚠️

Coverage data is based on head (f8beeff) compared to base (2a2c02c).
Patch coverage: 25.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #9083      +/-   ##
==========================================
- Coverage   92.97%   92.66%   -0.31%     
==========================================
  Files         459      459              
  Lines       37112    37115       +3     
  Branches     5365     5365              
==========================================
- Hits        34505    34393     -112     
- Misses       2080     2170      +90     
- Partials      527      552      +25     

Impacted Files	Coverage Δ
dvc/testing/fixtures.py	71.52% <25.00%> (-1.45%)	⬇️
tests/func/test_unprotect.py	78.57% <0.00%> (-21.43%)	⬇️
tests/unit/utils/test_utils.py	83.33% <0.00%> (-16.67%)	⬇️
tests/unit/test_daemon.py	85.18% <0.00%> (-14.82%)	⬇️
dvc/utils/fs.py	83.63% <0.00%> (-10.91%)	⬇️
tests/unit/proc/test_manager.py	89.28% <0.00%> (-10.72%)	⬇️
dvc/proc/manager.py	66.66% <0.00%> (-9.68%)	⬇️
tests/unit/utils/test_fs.py	88.42% <0.00%> (-9.48%)	⬇️
dvc/proc/exceptions.py	86.66% <0.00%> (-6.67%)	⬇️
dvc/config_schema.py	89.18% <0.00%> (-5.41%)	⬇️
... and 15 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[skshetry]

I don't think this PR does anything. The fixture is return via yield.

[skshetry]

This is also incompatible with pytest_docker>=1 as well.

[efiop]

I'll take a closer look at it...

[skshetry]

I tested dvc-ssh without this PR on both pytest_docker>=1 and pytest_docker<1, and it works. I think dvc_compose was the only thing that was making it fail.

[efiop]

@skshetry Hm, looks like I'm wrong here. Created a revert PR. dvc-ssh needs some investigation.