Fits a Negative Binomial model to the observation data and returns
the log-likelihood.
This is a wrapper around pypomp.benchmarks.negbin.
Parameters:
autoregressive (bool, optional) – If True, fits an AR(1) model.
Defaults to False (iid).
suppress_warnings (bool, optional) – If True, suppresses individual warnings from statsmodels/optimization
and issues a summary warning instead. Defaults to True.