Fix marking branch for removal after merge in Gitlab

[WouldYouKindly]

Hi.

I renamed should_remove_source_branch to remove_source_branch, and flag is now correctly set on merge request.

Tested on GitLab Community Edition 11.9.4

According to https://gitlab.com/gitlab-org/gitlab-ce/blob/master/doc/api/merge_requests.md#L745

[codecov]

Codecov Report

Merging #348 into master will increase coverage by 0.28%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #348      +/-   ##
==========================================
+ Coverage   93.92%   94.20%   +0.28%     
==========================================
  Files          11       11              
  Lines        1103     1157      +54     
==========================================
+ Hits         1036     1090      +54     
  Misses         67       67              

Impacted Files	Coverage Δ
pyup/providers/gitlab.py	85.24% <100.00%> (+0.37%)	⬆️
pyup/providers/github.py	99.37% <0.00%> (-0.01%)	⬇️
pyup/updates.py	100.00% <0.00%> (ø)
pyup/config.py	98.88% <0.00%> (+0.01%)	⬆️
pyup/bot.py	97.88% <0.00%> (+0.04%)	⬆️
pyup/pullrequest.py	97.22% <0.00%> (+0.44%)	⬆️
pyup/requirements.py	88.46% <0.00%> (+0.96%)	⬆️

---

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 b20fa88...1556368. Read the comment docs.

[rafaelpivato]

This looks to me. Unfortunately, as @htgoebel suggested, we should restrict the change-set to the actual GitLab call. We should have people relying on the actual old config values, and until we have means to be able to gracefully deprecate that, we need to keep supporting the old config.

Can you restrict the change-set to the GitLab call and any respective test whether pertinent?

[rafaelpivato]

Thanks for showing where the problem was, @WouldYouKindly 👍

[WouldYouKindly]

@rafaelpivato thanks for taking the time to fix it!