1) Original test with everything assembled showed 0Ω resistance across the MF/HF +/- terminals on the back of the speaker.
2) Those terminals connect directly to IP+ and IP- on the crossover PCB.
3) With the crossover disconnected the DC resistance between IP+ and IP- was open circuit.
If points 1-3 are correct and you’ve put it back together and measured 0Ω across the loudspeaker terminals then there’s a short somewhere between the terminals and where the wires connect to IP+ and IP-.
totally correct
Okay, that should narrow it down a bit and you can stop worrying about the crossover. If you only get the short when the loudspeaker is reassembled then you're looking for something between the crossover PCB and the loudspeaker inputs. If you disconnect the crossover you should be able to track it down... what are the binding posts on the loudspeaker mounted to, plastic or could it be anodized aluminium?