Add a function to suggest good value for the theta parameter

[Militeee]

As the choice of the library size distribution prior is somewhat hard if you are not familiar with Gamma distributions. A function that adapts the parameter to the data in an approximate way can be useful to standardize the inference

[Militeee]

Added function approx_theta_prior_params