Bruce Barlow did it with papers and paper developers. It turned into a pretty massive undertaking (I think he has posted about it in here). I've seen the resulting prints, and overall the differences are really subtle. In most cases it's a difference in feeling vs. something you can actually point to. Once you get past grain differences, the result for film would be similar, and probably more so.

I don't want to say such a test with B&W film isn't feasible, but it would be a huge challenge, and I don't think the result would yield much useful information except for the tester.

What is feasible, is to take two or three films of interest to you, shoot subjects that you like to shoot, process them the same, make some sample prints and see what you like. Then, take that film make some more shots and tweak the process and look at some more prints. If nothing else, once you've gone through that, you will have learned enough that a massive head to head test probably isn't necessary.