txt2jed has wrong bit ordering within word

[JohnDMcMaster]

Although it appears that the script correctly identifies which word the bit goes into, the position within the word appears to be wrong. This roughly lines up with me thinking the observed bit order was wrong. Unfortunately the microscope images also appear to be noisy, so this is a bit harder to quickly fix than I was hoping.

When i was looking through test scripts I found something that mixed up %X and %u on addresses. This is possibly related, but I would have guessed it would have generated wrong word order, not bit order.