[feature request] Support for zero inflated and hurdle models

[suntzuisafterU]

See https://www.rdocumentation.org/packages/pscl/versions/1.5.2/topics/zeroinfl for an example implementation in R.

These models are extremely useful in bioinformatics research. Large volumes of DNA sequencing data is being generated from projects such as:

• American Gut Project (AGP) (McDonald et al., 2018)
• Human Microbiome Project (HMP) (Huttenhower et al., 2012)
• Earth Microbiome Project (EMP) (Bahram et al., 2018)
• TARA Ocean project (Sunagawa et al., 2015) That can benefit from these models.