Reduce loop overruns to decrease robot response latency and reduce error logging

[jamesdooley4]

The logs and drive station dashboards have a continual output of CommandScheduler loop overrun errors that makes it hard to see anything else going on, and likely are a signal that there's robot responsiveness latency caused by the command loop frequently running over its allotted 20ms (0.02s)

[RobototesProgrammers]

From glancing at drive station console from time to time, this seems to mostly be from Shuffleboard.update(). Looking at logs would also be good to confirm this (and to see how long Shuffleboard.update() is taking).

[iamawesomecat]

where is this run? How can we make it cause less overruns? @jamesdooley4 @markpete