Instead of adding driver IDs by hand to a header file, generate the
driver ID from the (hopefully unique) name of the driver.
This means one file less for driver authors to modify, and one possible
source of merge conflicts less as well.
However, this means that already enrolled fingerprints will need to be
enrolled again, as the driver IDs will have changed compared to their
old on-disk value.
This will allow us to segragate helper functions that we want to use
within NBIS. The first helper will be a macro checking for overflow in
multiplications.
