Y-axis of clock plots should be shifted according to clock offsets

[martijnbastiaan]

It's currently hard to interpret the clock plots as they are relative to the measurement clock, which is (IMO) the main reason #604 has been able to go unnoticed for multiple months. I propose assuming one of the clock offsets is correct, and shifting the y-axis until it is set in its "proper" place.

E.g., this (example plots still affected by #606):

Should be approximately:

Should also be mentioned in #603

We should probably also be able to generate "debug" plots, where we show the position of the reference clock.