They tried to go for a bigger sample size, and only got a small one - happens. Suspiciously small for the size they originally tested (came to my mind as well), but thought it was worthwhile to say that they originally tested around 500, instead of you just focusing down on 37 - because it fits your argument.
Not optimal sperm parameters, but normal ones. Which - if they f.e. would work in a malaria ward - could be somewhat rationally explained (I'm no expert) perhaps. But agree that the 37 in the end are far from statistically representative - but as studies go, sometimes you take what you can get. Reading as if written by a child can be explained by the authors being foreigners, they use the terms of the trade though, so I'm inclined to believe, that you are embelishing your fact finding mojo here, same as with names -- sometimes the best studies come from where people having an interest in finding that information. So if Ivermectin is largely used at scale in developing countries, you wont find those studies from reknown institutions...
So agree on 37 actually was a red flag. But not one fit to denounce the entire thing, and that pretty much you go to town trying to say how prolific you are on all other points, ...
The actual red lines, where you probably should discard the findings are those, that the study apparently wasnt peer reviewed (otherwise reviewers would have corrected grammatical errors), and that the "journal" sold publishing space for a high service charge, and now apparently has vanished from twitter, where it was banned, and Snopes not being able to contact the study authors.
You dont dismiss a study, just because it only had 37 participants willing to jerk off into a cup. You just acknowledge, that it is not statistically representative. But you might dismiss the study based on all the other findings, hinting at "issues" with scientific method. (Of which not being representative is one - could still have been valid though, but then there are the other issues with it.)