I don't think it does; sure, someone could potentially craft a malicious commit that causes a SHA1 collision in your repo, but I think if you are merging commits from malicious authors, you've got way bigger problems than that.
Not to say that this attack is in any way practical, yet. Just that some providers don't require active involvement to try and attempt it.
As noted in the article, an SHA-1 collision attack does not appear practical now, but that is a situation that can change.