Mode 01: Master counter has glitch after 32 integrations

[gitj]

The current Mode 01 designs have a glitch in the master counter value (stored in the berkeley-vegas-fits field TIMECNTR) at the 32nd integration (when the acc_len is 768). That is, if you look at the pairwise differences of the TIMECNTR value, they are all the same except on the 32nd integration. This is not a show stopper, but should be looked into.

[gitj]

The number 32 is very likely related to the sync period which is currently set to once every 32 integrations. So it's likely that the first sync counter sync pulse is coming at a slightly wrong time.