A colleague is strongly opposing, and will not let me merge it until I find a customer that contacted the company because it's affected by this issue.
What can I do?
More seriously, working with people who don't care about what they're building enough to want it to be a better product is hard. I don't really want to recommend that you resign and find a new job, but two decades of experience writing web software tells me that's probably the only real solution. If you're on a team with people who don't actively want the code to be as good as possible, and who are willing to put the effort in to making it that way, will be bad for your mental health and possibly bad for your career if you pick up that habit.
Tossing aside the accessibility angle, fly by night changes often have ramifications that the authors did not anticipate. I have and would again veto change requests that not part of our planning process and offered as: "Surprise, here is this cool thing I did."
other developers reviewed the change and have been tested to be sure we don't have regressions. The company is small, we are only 6 frontend devs.
You see that other devs have already reviewed the changes as a good thing. In my opinion this is the kiss of death for your case. Not only are you working on un-prioritized work but now others are as well.
Work on stuff that has been prioritized. Don't get side tracked on work that has not been prioritized. If you believe it is important work, submit a bug report or feature request and get it prioritized. Then you can legit fix it and get it folded in.
I know it seems petty, and I have been guilty of 'drive by refactoring" when I was a junior dev so I understand the temptation to "fix things".
This is the type of issues you want to fix before are reported, not after.
It's like waiting for someone to die before adding a zebra crossing.
If it was/is that important then you can create a ticket right?