Oddities with time sync collections

[yarikoptic]

As observed in analysis by @vmdocua , this is a complement to

• Synchronization (or tracking of offset) of involved clocks #13

based on new recordings from https://github.com/ReproNim/reproflow-data-sync/blob/4e58126cbe7429f5d1ba9973d4f4a87cdb64c5d6/ses-20240809

1. on birch, it doesn't flip 8th bit for triggers some times. Emailed Ben with details (subject 'alink_byte and strobe_bit oddity'). More on that in https://github.com/ReproNim/reproflow-data-sync/blob/nf-timing-sync1/code/README.md#notes-and-observations
2. reproevents (micropython) client_time clock slows down 10x or more for a bit and then goes on normally. Emailed @TheChymera with details (subject 'reproevents client time oddity'). Related issue:
   Tune up reproevents timing data in micropython/client #110
3. Oddity of "jumping offset" of DICOM's AcquisitionTime across sequences #109