I noticed something about this challenge. The documentation says that for this model, there are only 960 unique substitution alphabets, but with the wheel settings given I generated a full 1024 different alphabets. I did it with two different simulators, one of which was the JAR file in the challenge. (This does not affect my solution.)
So I looked deeper into Weierud's documentation, and what is happening is this: For various permutation-circuit wirings, there are different numbers of possible permutations that can result. The Germans accepted only those that had 30 or 32 possibilities, and so there were 30x32=960 or 32x32=1024 different alphabets. Other wirings were rejected. So when the documents that come with the challenge say "360 possible" wirings, it really means "360 acceptable" wirings.
I made a list of the 360 acceptable wirings, but since the list is 360 lines long, I am not posting it here.