Problem is the temperature keeps changing (upwards). If it was a step input, then you might see a sudden die-off and then a ramp back up. But the temperature change is, to first order, a ramp input.
I'm willing to bet that the necessary genes already exist in the population, since the bugs already handle daily temperature swings that are much larger than the increases from climate change.
Given that, a few dozen generations is plenty. And insect generations are for the most part quite short.
It has, apparently, been going on for a couple of decades now. As you correctly point out, this constitutes many insect generations. If this were simple case of letting pre-existing genetic variations that are robust to temperature get selected for, we would be seeing populations rebound.
And it seems most of the other factors, besides climate, have been accounted for. The losses here are in large nature preserves, where insecticides and herbicides are not used. And in the case of Puerto Rico, at least, pesticide use island-wide has dropped dramatically since the 1960s. So one would not expect the results of pesticide use on insect populations to only be manifesting themselves now. Other than climate, most other factors seem to have changed in the insects' favor. Yes, neonicotinoids are an issue, but a relatively local one. One would not expect them to be an issue deep inside a Puerto Rican rainforest nature preserve.