Update on the SBM prior: calculating the number of cluster and the allocation of the nodes

[sekulovskin]

The following changes have been made:

• We now save the cluster allocation vectors from the Gibbs sampler
• When save = FALSE we use the cluster allocation vector to calculate the cluster membership (using Dahl's method), as well as the posterior probability of the number of clusters. For the latter, I decided not to use Equation 3.7 from Miller and Harrsion, but instead simply use the sampled allocation vectors (see the summary_SBM function in the utility_functions script for more details).
• When save = TRUE, we simply return a matrix of the sampled allocation vectors in each row. Users can then manually make use of the summary_SBM function.
• I also added some additional clarifications in the roxygen code.

[MaartenMarsman]

Looking good!