I've read the full thread and have thought about it for a lot of the day. I 100% agree that achievements should not be criteria for getting a game published or algorithmically boosted, but I think that there's a mix of feeling coming out of you hating achievements, and bad achievement design.
I don't believe achievement design is a zero sum game where lots of people are always annoyed.
I'd recommend the Gemsbok's video on achievement design, It might give you more idea of why people value achievements, such as wanting to pay tribute to your game: https://www.youtube.com/watch?v=ts3X9jSlcRE
The example you gave of using achievements to gamify turning on a menu setting is bad because it's just a clumsy way to signpost a settings option, and you're 1000% right that it should be done a different way. There's even a joke one about this in The Stanley Parable UD: 'Settings World Champion'
When achievements are an afterthought, I'd recommend doing them in the simplest way possible, main progression achievements with little to no unique artwork, just use character sprites or something pre-existing. When not, I do think there's a world out there for both good and bad achievement design.
Because you mentioned, limited time achievements prevent commitment to 100%. Ones that are too hard can also prevent this, but they're more intentional. Some players will be annoyed because they want to pay this 100% tribute, creating dissonance between how they feel and how they want to show it.
They're not evil gamer goblins who hate you, (ok.. some are... achievement number hoarders do exist) but for most, this is the exact same design problem as when normal completion of a game's story is too hard for some players that still want to play it, but you can't cater to everyone.
Comments
I'd recommend the Gemsbok's video on achievement design, It might give you more idea of why people value achievements, such as wanting to pay tribute to your game: https://www.youtube.com/watch?v=ts3X9jSlcRE