Comment by vintermann

Everyone wants to drag nontermination into this, but in the OP's example, the compiler already had proof that a the conditional would never evaluate to true. What you can or can't prove in the bigger picture isn't so interesting when we already have the proof we need right now.

It's just that it used this proof to remove the conditional evaluation (and the branch) instead of warning the user that he was making a nonsensical if statement.

So to the question of "when can we hope to do it" the answer is, "not in all cases, sure, but certainly in this case".