If the fuse blows, it must mean too much current, so where is the problem, controller or motor?
Get a flashlight and make a visual inspection of the brush contacts and the contacts rotating under them. You look for pitted, irregular, very blackened, missing, splashes of solder, or eroded commutating contacts. You can turn the motor by hand. You can hear and feel any irregularities that suggest a damaged motor armature.
It is very possible that a capacitor snubber across the motor drive output connection or across the voltage power switch FETs is open and causing the failure. A scope can show excessive spikes that could blow a power FET. In series with a snubber capacitor will usually be a small wattage resistor that is not uncommon to have opened circuited.
Testing the output with a resistive load like light bulbs will not stress the controller with inductive kicks and surges. Testing it with another motor a similar size and type would be a better test, if you are lucky enough to find one.
You could also have a motor with just one wire shorted to the next, just two wires touching in a winding that is short circuiting the output or else a shorted commutator contacts(solder or accumulated carbon and/or copper dust) can short adjacent contacts on the rotating armature under the brushes) causing current surges in excess of normal at low speed settings and quickly blowing the fuse at highest speed settings.
Unless you can scope the current into the motor with a current probe, you might find this problem is difficult to troubleshoot, and finding a motor problem does provide a method to fix it, it might be easier to swap another motor.
Use any scope to probe into this problem:
Make a simple current transformer by winding just a few turns of wire upon a ferrite or metal core and next to it or over it, a secondary of several turns of wire. The secondary of this makeshift transformer must be, for safety reasons, made of insulated wire(but you only need a few turns) and so now you have an output isolated from the AC mains and this is your current sniffer winding fed to a scope. Then you connect the primary of this transformer in series with one of the two power leads going to your motor. Crank up the speed to the motor slowly and you can see how the motor behaves in commutation of power. The waveform may be full or spikes, but should be consistent. A large difference in current will show up as a big spike during every armature rotation cycle and tell you where the problem is. If it is the motor, the trouble will be be clearly visible on the scope as a surge.