I tend to be in the camp of "scan everything with an electron microscope and simulate lasers", but honestly there is a point where a decision has to be made.
Now, it's important to be _sure_ there isn't any useful data, even just silly intentional data, in that space… but from what I've learnt reading around, it would seem that its genuinely "useless" and unused. As long as the beginning and end of these sectors are well known are documented, I see no harm with the current method which ignores those bits.