There have been rules changes to try to address this and other bench press issues, the main ones being heels on the floor and head on the bench (rules the IPF introduced several years ago but others haven't followed). But even with this, it is possible for super flexible people to still maintain a huge arch. Banning heeled shoes in bench press would also help (assuming you have to keep your heels on the floor).
Some factions in the IPF (mainly from USAPL) tried to introduce a rule that would require that the lower leg could not be more than 90 degrees to the ground. This resolution did not pass. It's not a bad idea as it would still allow some degree of arching but not a ridiculous amount. It is directly addresses the risk of injury, as the lumbar hyperextension can only happen if you are able to tuck your feet under the hips. The problem is in the judging.
The other idea was to impose a minimum stroke length. This also failed to pass the IPF congress. Frankly, I can see why. It's a stupid rule. If you've got ridiculous short arms, good luck to you, and you'll also have a distinct disadvantage in the deadlift that tends to balance things out.
Beyond this, it is difficult to think of a rule that can actually address arched benching. You could have a rule that specified that the entire back is in contact with the bench the entire lift. A similar rule exists for the glutes - they must stay on the bench. But this would actually make bench pressing more dangerous for the majority of lifters, ie those that don't have a ridiculous arch. Some degree of arching definitely makes the bench press a safer exercise. From an aesthetic point of view, the problem isn't arching itself, it is the degree of arching that seems to cause people concern. The first problem you then face is where do you draw the line between an acceptable and unacceptable degree of arching? Secondly, how do you express a rule that actually allows referee to enforce this limit?