apply self-calib to precondition prior

[hyunjimoon]

We need prior and posterior to be similar ranges for valid testing (e.g. power). For two parameter normal models, tail area of scale papermeter is honed after over twenty iteration.

Further improvement:

1. all parameters are updated 1d way, but 2d (loc, scale joint) is possible with 2d binning.
2. distance-based metropolis hastings explained here will be stashed for now as it can be

• weak to simulation noise (e.g. possible to early stopping by chance)
• need theoretical justification such as reversibility, detailed balance etc.