Realistically, extermination of males is exactly what it means. It cannot be stochastic drift; the pool of chromosomes dying out is too large. It can be selection (it is selection), but it can only be the kind of selection that involves every native male failing to have children at all, because the Y chromosome does not recombine.

First of all, given that we only have a limited sample of the Y-chromosome space, I don't think we can conclude that no native Y-chromosomes actually survived. That aside, it's easy to get selection for specific Y chromosomes without extermination, for example if you have a ruling caste whose males are favored (e.g. by primogeniture) or through preferential mating by lineage. See: the prevalence of the (presumed) Chingisid chromosome here: https://doi.org/10.1086%2F367774