I don't see how this would be a long term solution. The bacteriophages would evolve to expand, or even switch their host selection pretty quickly if similar non

I don't see how this would be a long term solution. The bacteriophages would evolve to expand, or even switch their host selection pretty quickly if similar non-resistant strains of E. coli are available. E. coli is a pretty fast adapter itself. It'd become an evolutionary arms race, and as we've seen from the Lenski experiments, that does not result in the destruction of the pathogen's host.