I agree that deciding between MWI and other interpretations seems impossible, so it's not a problem that's considered important in physics, more of a good topic

I agree that deciding between MWI and other interpretations seems impossible, so it's not a problem that's considered important in physics, more of a good topic for a dinner conversation among physicists. Our understanding of quantum measurements is much better now than it was at the advent of quantum mechanics, and it's understood that there's no such thing as an "abrupt" collapse of a wave function during measurement, it's instead a continous process involving entanglement and decoherence between two quantum systems (the system being measured and the measurement system). This understanding allows us to explain any measurement process using only "plain" quantum mechanics, i.e. unitary evolution of two coupled systems described by a Hamiltonian. If we just look at the Hamiltonian of the measured system it will seem that it shows wave function collapse because its evolution is non-unitary, but the evolution of the entire system is still unitary. From this angle I personally find any further invokation of a wave-function collapse mechanism unnecessary, and I think it overcomplicates quantum mechanics without giving us any benefits, so we might as well do away with it. Unfortunately, most courses on quantum mechanics still introduce the students to wave-function collapse when explaining measurement outcomes, which might be the reason that so many physicists have a bad understanding of the measurement process. Of course wave-function collapse is still a good shortctut for describing strong quantum measurements, I think it should be introduced as an approximation of a continuous measurement though.