Reduction in population allocates more reproduction-relevant resources to the survivors. Resistance genes spread faster if the threat kills the non-resistant. (It is much easier to outcompete a corpse than an invalid.)
But ancestor post is also presuming that it is possible for bees to acquire genetic resistance to CCD. As we are not entirely certain about the exact mechanism involved, which may or may not have something to do with pesticides, chemicals in HFCS, dietary deficiency disease, or mites, it's hard to say whether any random mutation could protect against CCD without also making the colony unable to reproduce. For instance, if CCD is caused by a dietary deficiency, a colony is very unlikely to spontaneously acquire the ability to synthesize the missing nutrient, unless the species lost that ability at some point.
For instance, it is far more likely that humans could re-acquire the ability to synthesize ascorbate than it is that we might spontaneously gain the ability to synthesize our own omega-3 fatty acids. E. coli evolution experiments show it is possible, but the selection pressure has to be maintained over thousands of generations.
Applied human brainpower to the specific mechanisms of the threat is going to cure CCD much faster than bee breeding, which in turn is faster than natural bee evolution.
The only good behind the pruning of the population is that it is scaring the human agricultural industry into investing resources to stop further losses.