test for method brglm.fit

[guillermoaguilar]

we allow glm.fit and brglm.fit

so far only glm.fit is tested.

brglm.fit should be also tested

[guillermoaguilar]

and explore other ways of regularizing the parameters.
Because sometimes the optimization gives numbers way too high, e.g. 3000

[JorisVincent]

Looked into this briefly -- this not a feature of the R {{MLCM}} package, correct? Instead, it looks like we would have to include quite a bit of our own R code.

Alternatively, we could try to integrate this into {{MLCM}} itself...