This is in reference to a question recently posted by the user Prajneya which involves a Linguistics Olympiad puzzle. Since I have been into the Olympiad for two years now, at least 500-600 distinct questions exist and if people ask for solutions to solving a problem isn't it better suited for puzzling SE ?
It's not clearly off-topic, but it's not clearly on, either. Part of the problem is that we have stipulated that requests for trees are off topic, without clearly identifying why they are OT. My own rationalization of that ban is that it's a version of the principle "We don't do your homework for you". As far as I can tell, though, these competitions make people sit in room with a proctor, so these are practice problems, not "real" problems.
Since the methodology of linguistic analysis is utterly central to linguistics, then I maintain that methodological questions are on topic. A good methodological question would actually identify something about methodology and frame the question in terms of a methodological principle. I grant, though, that that sets the bar pretty high, since a lot of analysis out there is unprincipled.
A question of the form "Here is my analysis: did I screw it up?" can have good answers if the person screwed it up, but not if they didn't. So I don't see any clear answer to the question, unless you look at the help center and see that indeed such question are literally not off topic, but people come up with custom-close reasons anyhow.
I don't know if there is a premise that LO questions should be answerable without any knowledge of linguistics. If that is a basic rule, such questions might be more on topic for a puzzle-solving SE (I take it there is one). But I don't think the sheer volume of potential questions makes the questions non-linguistic. On the other hand, since the questions that I have seen are merely exercises in combinatorics, there is basically just one good question about such problems, and then the remainder are duplicates of that one good question. So maybe you could go through a half-dozen problems and extract the One Great Question that covers all LO questions, ask it, and we'll have a defense against the onslaught.