Knowing the difficulty of a question can be counter-productive. For example, when seeing a hard problem, you subconsciously start looking at it differently when in reality its solution might be much easier than you expect it to be. Similarly, for medium problems, you might perceive a problem to be much easier than it actually is.

Understanding the constraints help in addressing these issues, but this does not translate to real-life. In interviews, you do not know whether a problem is a 'medium' or a 'hard', or about the exact constraints. Depending on how your brain is wired in problem solving, all your practice might very well mean nothing in such situations.

This is similar to chess. It is a well-known hypothesis that one can solve really tough positions doing chess puzzles since they are aware of the fact that there is always a winning move. But during actual games, this doesn't translate well since you don't have this information of whether or not you are in an advantageous position.