There are a couple of problems with this line of reasoning - yes, as an RNA virus, there is a high rate of mutation. However because of the redundancy of amino acid encoding most mutations will be preservative (ie produce the same protein). Additionally, although there is no reason that chance can’t produce the exact same spike protein mutation in another location in the world, it will be of a different lineage which will clearly show up on sequencing as there will be some clear differences elsewhere (in the preservative mutations). This is highly unlikely to be a result of immune response pressure, in fact I would discount this entirely, mutations arise as a response to copying errors and inside an infection cycle (initial infection, replication, transmission, adaptive immune response).

The final immune response will shut down the virus ability to replicate freely due to recognition; selective pressure to avoid immune response is unlikely due to 1) the continually lowering viral load and 2) the very high number of non-selective antibodies produced that recognise a large number of viral epitopes.