Add data set for text analysis

[ablaom]

Taken from the MLJText.jl requirements for transformers:

Generate a vector whose elements are either tokenized documents or bags of words/ngrams. Specifically, each element would be one of the following:

• A vector of abstract strings (tokens), e.g., ["I", "like", "Sam",
  ".", "Sam", "is", "nice", "."] (scitype AbstractVector{Textual})

• A dictionary of counts, indexed on abstract strings, e.g.,
  Dict("I"=>1, "Sam"=>2, "Sam is"=>1) (scitype Multiset{Textual}})

• A dictionary of counts, indexed on plain ngrams, e.g.,
  Dict(("I",)=>1, ("Sam",)=>2, ("I", "Sam")=>1) (scitype
  Multiset{<:NTuple{N,Textual} where N}); here a plain ngram is a
  tuple of abstract strings.