Core `gen_additive_mod()` algorithm

Develop `gen_additive_mod()` algorithm with modes "regression" and "classification"

- Use `parsnip::linear_reg()` as an example.
- Linear Reg: https://github.com/tidymodels/parsnip/blob/master/R/linear_reg.R
- Linear Reg Data: https://github.com/tidymodels/parsnip/blob/master/R/linear_reg_data.R
- Use `multilevelmod::linear_reg()` as an example:
- Linear Reg Data: They are adding new `linear_reg()` engines: https://github.com/tidymodels/multilevelmod/blob/master/R/linear_reg_data.R
- Test with `parsnip` and `workflows` interfaces

See business-science/modeltime#71 for Discussion and Basic Example

Read the referenced parsnip R files, especially linear_reg.R and linear_reg_data.R, then compare the multilevelmod linear_reg_data.R example. Use the parsnip and workflows interfaces to test gen_additive_mod() in both regression and classification modes; done means the core algorithm supports those modes and works through both interfaces.

Written by the indexing model from the issue text.