Changes in version 1.1.2

o Improved numerical stability of Hurdle-NB parameter estimation.

o Added safeguards against non-finite likelihood values during
  BFGS optimization.

o Improved handling of extreme negative binomial parameters and
  zero-truncation probabilities.

o Added protection against zero or non-finite initial parameter
  values.

o Improved initialization of group-specific mean parameters in
  tag-wise dispersion estimation.

o Added probability clipping to prevent infinite logit
  transformations.

o Improved robustness of tag-wise dispersion estimation for sparse
  genes and genes with all-zero counts.