Here's my theory of how it may have happened. Stage 0: 5 million years ago they were the same species. Stage 1: Subspecies. Ibericus and Structor became distinct populations of the same species still capable of mating. Stage 2: Parasitism. Structur became capable of replacing Ibericus dna with their own in eggs. Stage 3: Loss of function. Because of rampant structor parasitism nearly all workers were structor. So when Ibericus lost the ability to make their own workers it was a neutral mutation.

If I understand it correctly their theory is as follows. Stage 0: 5 million years ago they were the same species. Stage 1: Subspecies. Ibericus and Structor became distinct populations of the same species still capable of mating. Stage 2. Loss of function. Ibericus lost the ability to create their own workers, but as Ibericus and Structor existed in the same places hybrid workers allowed Ibericus to survive despite this. Stage 3: Ibericus learns to clone structor males to live in places where there are no Structors naturally.

Kinda interesting that even though the end result is the same who is considered the parasite is different.

One of the author of the study here. Your scenario is very good. The reason why we believe that the second one is way more likely is because there are several other species of the Messor genus that need to hybridize to produce workers, so it's very likely that it's the ancestral state and that male cloning comes after.

Note that this second scenario doesn't necessarily imply that M. structor males are not some sort of parasites. Actually, because it's detrimental to M. structor males to father only workers (they don't transmit their genes to the next generation), it's easy to imagine selection to remove DNA of the mother so they can be cloned in a fertile caste (males). That being said, maternal DNA elimination prompted by the male genome would be nearly immediately also beneficial to M. ibericus (because having M. structor males at home is good for them), so this maternal DNA removing is something that benefits to both partners, which is why it's not that difficult to imagine that it evolved and has been succesfully selected. When something benefits to both partners of any relationship, there are good chances that it will be selected at some point.