Transpose: address inadequate naming

[wvdvegte]

Similar to what was reported in #7084 regarding Feature Statistics, the output of Transpose assigns the header Feature Name to its first column while it doesn't necessarily hold just features, but also metas and targets. A quick fix would be to rename this column to Variable Name or Variable.

Actually, I propose some further improvements, since Transpose is particular useful for datasets like this one:

The screenshot shows a set of users with ratings they gave to book titles. At some point, one might be interested in finding similar users based on distances between their rating profiles. But at some other point, in the same workflow (or another one), one might be interested in finding similar books based on users that gave them similar ratings. For the second investigation, Transpose can be applied. In that case, it would be nice to have an option to user-define the name of the first column.

[janezd]

@markotoplak -- who is otherwise a big proponent of "Column" -- warned that if we change the name of the column in the output table, this may break some existing workflows, e.g. if this name is used in Select columns.

We can make the change so that existing workflows will keep using the word "Feature", and new workflows will use "Column", yet this complicates the widget code.

So the idea now is to make this change if keeping compatibility is a minor nuisance, otherwise we keep it as it is.

[wvdvegte]

I guess it is a minor nuisance if Transpose isn't used very often, which I think is the case ...