It seems to be a PTC thermistor, as the resistance increased when I heated the hub rather than decreasing
Unfortunately the slope and offset can vary and it's not linear
The reading at ambient (22 Deg C 72F) is 584 ohms
It isn't a thermocouple on the MXUS and both wires are black from your display so that isn't one either so you have half a chance of getting a reading but a good chance it may be wrong. You may also find it only registers at all when the motor is hot but it may be useable if you know how the display relates to the actual temperature (use brain for calibration)
I might suggest double checking that there is low resistance between the throttle wire supply and the red wire that goes to the hall sensors? (check for a low resistance between the two unpowered)
If it's low resistance it 's a common 5V rather than an isolated one (sometimes used to avoid interference)
The other thing to be sure of is which pole is positive coming out of the display. both wires on the existing (axle) probe are black because it doesn't matter but it will only work one way connecting to the hub sensor.
With the new (axle) sensor powered up, check the voltage to each wire of the sensor.
The white wire from the MXUS should connect to the low voltage side of your new axle sensor display
Peer review / sanity check - it's a bit late here
I tried short circuiting my MUXUS to heat it but as it's 0.22 ohm resistnance my benchtop PSU did nothing feeding it 10W.
A hot air gun got a rise from 573 to 584 ohm but who knows what the coil temperatures were
Couldn't fit it in the oven. Partner will not be amused if it falls of the hot water tank in the middle of the night so working on other methods of calibration