Fix datarace on git.GlobalCommandArgs on tests

[lunny]

Should be part of #1441

[guillep2k]

If the reason to constantly change the global options is that some commands require LFS settings and some commands don't, isn't it better to simply have two sets of global options?

[zeripath]

@guillep2k so in the git_test we create some LFS files as a client. We want the LFS filters to run to install them into the client repository that we are going to push to Gitea. That is the only reason for changing the filters.

[lunny]

@guillep2k Only a little git commands need that changes, so I think that's unnecessary.

[zeripath]

Apart from name suggestion this lgtm

[codecov-io]

Codecov Report

Merging #9162 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #9162      +/-   ##
==========================================
+ Coverage   41.01%   41.03%   +0.01%     
==========================================
  Files         553      553              
  Lines       72026    72041      +15     
==========================================
+ Hits        29545    29562      +17     
+ Misses      38750    38747       -3     
- Partials     3731     3732       +1

Impacted Files	Coverage Δ
modules/git/command.go	92% <100%> (+0.42%)	⬆️
modules/git/repo.go	45.11% <100%> (+1.04%)	⬆️
modules/git/commit.go	52.09% <100%> (+0.94%)	⬆️
models/unit.go	62.16% <0%> (-5.41%)	⬇️
modules/indexer/indexer.go	55.26% <0%> (+10.52%)	⬆️

---

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 055f6d2...1826547. Read the comment docs.

[sapk]

make l-gtm work