The control contacts in overload relays were never particularly robust. I'm sure this had something to do with the NEMA standard. Maybe newer OL relays are better.
I think NEMA went to the trouble of having a separate paragraph to explain this because they agree it is *usually* best not to be switching on the neutral side. Most remote contacts are open/closed more frequently than the OL contacts, so contact welding would be more likely to be detected. Also, in many cases, the length of control circuit wiring to field devices would greatly reduce the short circuit current.
I worked (years ago) for a large consulting firm that always specified the overload relay contacts to be on the line side, and it was never a problem to get them built this way. But that was before the time that MCCs were UL-labeled. I don't know if the MCC could be UL-labeled if the OL contact wiring was not per NEMA standards, because I suspect that is the way all the testing was done. Maybe someone else knows the answer to that.
But my main point was that this has always been somewhat controversial and that NEMA has a stated engineering reason for their standard. I suspect the wording in NEC Article 430 was carefully crafted specifically to allow this practice.