By the time the code is cleaned up, I offer advice on how someone else might implement it, unless I find something deeply philosophically engaging about the feature proposal, like thinking of something even better than what they asked for.
Getting mad at people who ask you to fix your own mess is a defense mechanism of people whose egos have outrun their abilities. I work with someone like that (and an enabler) and I will veto him ever working with me again. It’s a special kind of toxicity that feeds other people’s impostor syndrome.
My suggestions to fellow developers; Every bug, feature request and question (good FAQ candidates) is the contributions, if you are not interested in accepting, just be upfront and polite.
People ask questions when they are honestly trying to figure out what they need that they aren’t getting, and because you have experience, you can often see where they are headed (a vector value, if you will) “Out of the mouths of babes” comes wisdom that only you appreciate. Questions are, to me, the highest quality critiques I ever get, and the magnitude is relative to the number of people who ask me the same question.
I think this changes the problem significantly. A feature request is essentially asking people to put in effort for you; that shouldn't happen if you have other options. If you're in a position to build something yourself, or to pay someone to deliver a feature you need, then that should be your first choice. Alternatively, you could propose sponsoring the feature eg paying the project owners to build it.
If the project owners are open to a merging the feature in to the project once it's complete that's great, everyone benefits from your effort/money, but requesting a feature that you could build yourself feels a bit wrong to me.