Atleast for the example question I feel like it's like one of those stupid interview questions. It's a perfectly valid approach to accept the problem as it is - the clear implication is "here's the problem, your mechanism for fixing the problem is to add bricks at X cost, go fix". If you want to subtract fine, but there's a shed load of things you need to consider - why was the pillar there in the first place, is the base structurally sound, etc. Again with the white/green square thing, people probably simply read the white as "unset" and green as "set" so they don't think to "unset" something becaues it may not be an obvious option - and, this is important, it's a trivial test they aren't invested in.

At the very least I would expect a symmetric experiment where they imply you should be removing bricks and can't add.