Implement basic bayesian method

[xflouris]

Implement a Bayesian approach to estimating the prior distribution using a uniform hyperprior

We have decided to focus on a very basic bayesian space exploration for now as described in many textbooks:

1. Generate an initial delimitation.
2. throw a coin to decide whether we are (a) turning a speciation node to a coalescent or (b) a coalescent root to a speciation.
3. Select a node from either (a) the list of speciation nodes or (b) the list of coalescent roots
4. Change it to a speciation or coalescent node, and compute the new likelihood
5. Based on the Metropolis-Hastings ratio compute the probability of accepting this new delimitation
6. generate a random number between 0 and 1 to decide whether to accept this new delimitation
7. repeat from 2

We will implement both version (multi and single)

[xflouris]

finished.