Location based extrapolation to reduce clock state spaces

[seblund]

This requires a static analysis of components akin to that in the paper on location based extrapolation in UPPAAL with a pretty clear motivation:

"So far, the maximum constants have been determined by a global examination of all guards in the model. In this paper we propose a coarser location-based abstraction, using location-dependent (and
smaller) maximum constants based on the relevant guards for a particular state of the timed automaton"

In the best case exponentially reduces the state spaces of many verification checks in the engine.