This is the next thing; first they created this elaborate backbone only to 'mutate' it away? You quite misunderstood Baric here; he used this consensus sequence to find a 'stable' genetic backbone that would work so he could study it in a lab. Because, and here is the point, if this bat derived backbone 'works' in human cell culture and or hACE-2 mice, it is easier to figure out which parts of the bat viral genome versus SARS viral genome confers that 'human' infectivity, and which prohibit it.

Furthermore, your suggestion that 500+ mutations that the consensus backbone is different from SARS-CoV-2 happened through some evolutionary mechanisms (like serial passage) after the original backbone was assembled is just laughable. A process like that would take years in a lab setting, and again, why would they do it to the backbone (and all over it in irrelevant places) when what they were really interested in was 'optimizing' the ACE2 affinity of the RBD?

Even if they were really interested in getting their consensus batSARSr backbone more infectious and adapted to humans, the faster way would be to just exchange and test different RBDs or spikes on a stable backbone, rather than go the evolution route at this point, really wasting time and energy and resources to observe how hundreds of mutations in unimportant regions evolve; all of them would of course then have to be tested for relevance and function all over again to really understand whether that mutation was just a random passaging artefact or actually provided an evolutionary advantage... it's just a no-no.