Then pretend it was a more interesting trap, like Grimtooth's lava trap or something. That's not the point--the point is to find a way to allow either players or character skills to engage with the same content. If you're using Grimtooth's The Bigger They Are trap, wouldn't it be a pity if it just turned into "I roll Investigation. Oh, it's a falling rock"? And it would also be a pity if maxed out passive Investigation 32 provided no benefit even against simple, obvious traps. If a player "pays" character resources to avoid certain kinds of content, he should get the benefit of his choice. Choosing the DCs in a range where normal humans would not detect the clues, but an optimized PC could detect the clues, seems like a pretty good compromise. Do you disagree?
Yes. Most traps (and especially the rotten floor you postulate) aren't that hard to detect. They're successful because they're somewhere or something that doesn't get looked at closely and/or are things that can't be avoided (a large pit trap that can't be jumped across by normals blocking a narrow hall, for instance). Your method is placing traps on the level of "normal people will always fall for them unless they pick the right pixel first." Most traps should be able to be found by a normal actually looking for them. I think the method of artificially raising DCs so that passive skills fail to detect while simultaneously having the right questions asked generate autosuccesses to be a poor design decision. If, for example, I was playing a character with a high passive perception and had no clue that the floor was rotten aside from a general statement of 'there's a faint smell of rot in the room, you can't tell where it's coming from' while the person who said, 'I try to tell where the smell is coming from' gets to know immediately, that's a failure in description to take into account the fact that my character is better at noticing such things. You've arbitrarily decided to remove agency from my character to grant it to another. Not a good trade-off.
Instead, the rotten floor can be easily noticed (I even advocate for blatant signalling, like with a previously collapsed section), but can't be 'disarmed'. It has to be navigated. That invites the players to engage the environment to figure out how to get from A to B across the rotten section of floor, and to ask questions, without inventing a pixel hunt or arbitrarily assigning DCs at ridiculous levels. Make traps that require player decisions to overcome, and not just rolls, and you've done exactly what you're looking for. And it's much better than invalidating player choices by arbitrary DC assignations or pixel hunts. If the players feel like the choices they've already made retain use (high perception) while still being able to make decisions that feel like they affect the world (engage the challenge to cross the floor with good understanding of the situation so their actions make sense in the world) rather than dismissing previous choices and requiring specific questions to determine new information, they'll appreciate the game more.
Also, I dislike the 'click' mechanic. The player has already been caught in a trap and now must guess the best immediate reaction to avoid the effects? If you telegraph heavily, I could get behind that, otherwise it's really just asking the player to gamble on a choice. Some players will be good at this, through experience or intelligence, and others will be less good at this. If I suck at synthesizing information in stressful situations, but I play a rogue with high intelligence and catlike reflexes, this method punishes me, the player. Vicely, if I'm good at making decisions quickly and synthesizing the situation (or just guessing) but my character is an ironbound lug, this system benefits me. I dislike systems that test players, not characters, on principle, and that's exactly what this does.
In fact, your whole approach here is one that favors testing players -- can they ask the right questions? Do they guess well when I yell 'click' at them? I dislike that. There's always some testing of players in a game, but mechanics that spotlight player choices instead of character choices aren't my cup of tea.