Feature Request: @unnest_wider kwargs prefix and deep

[lincolnhannah]

Request - Add optional keyword arguments to @unnest_wider

prefix (defaults to true). false removes the parent column prefix (except when a column name already exists in the Dataframe).
deep (defaults to false). true applies @unnest_wider to any sub NamedTuples, Dicts or Structs.

Example

struct X1
    C::Int64
    D::Int64
end

struct X2
    A::X1
    B::Int64   
end

v  = [ X2(X1(1,2),3), X2(X1(1,2),3) ]
df = DataFrame( v = [ X2(X1(1,2),3), X2(X1(1,2),3) ] )


@chain df begin         # Currenty Functionality 

   @unnest_wider v
   @unnest_wider v_A

end


@chain df begin         # Proposal 

    @unnest_wider v deep=true prefix=false
 
    #result   DataFrame( C=[1,1], D=[2,2], B=[3,3] )

end

Request 2
applying @select to a Vector of identical Structs or a StructVector is equivalent to appling:
DataFrame
@unnest_wider deep=true prefix=false
selecting colums.

@chain v begin        # Current 

    DataFrame
    @unnest_wider A

end


@chain v begin        # Proposal 
    
    @select  C D B
 
    #result   DataFrame( C=[1,1], D=[2,2], B=[3,3] )

end

Request 3
Extend to KeyedArray

using AxisKeys
ka = KeyedArray(   v,  id = ["k1","k2"]  )

@chain ka begin     # Current 
    
    DataFrame
    @unnest_wider value
    @unnest_wider value_A

end

@chain ka begin     # Proposal 

    @select id C D B
    
    #result   DataFrame( id=["k1","k2"], C=[1,1], D=[2,2], B=[3,3] )

end