Eliminate ml.estimation.trunc.dist dependency on the "family" argument

[wleoncio]

Motivation

The ml.estimation.trunc.dist() function requires a family argument to manually trigger the appropriate estimation method. As of commit 8e17b33, this is not required anymore, because the y argument already contains this information in its class.

Steps

• Remove family argument
• If using S3: change name of ml.estimation.trunc.dist function to something without . (name can be kept for S4)
• Split generic and methods for each signature
• Adjust tests and examples

[wleoncio]

On further inspection, this might not even be necessary, given how the cases for family = seem to only contain functions that could be made into generics. Namely:

• init.parms (issue Make other functions called by ml.estimation.trunc.dist #21)
• sufficient.T (issue Make other functions called by ml.estimation.trunc.dist #21)
• average.T (issue Make other functions called by ml.estimation.trunc.dist #21)
• natural2parameters (issue Export natural2parameters and parameters2natural #9)
• parameters2natural (issue Export natural2parameters and parameters2natural #9)
• density.trunc (issue Export density* functions as S3 methods of dtrunc #5)
• get.grad.E.T.inv (issue Make other functions called by ml.estimation.trunc.dist #21)
• get.y.seq (issue Make other functions called by ml.estimation.trunc.dist #21)

So if all those are made into generic functions, I think we can simply eliminate the family checks and leave the rest as is.

[wleoncio]

Points 2 and 3 from OP can be addressed later on their respective issue numbers.