The function considers the following properties:
the maximum number of characters (to identify free text fields with long entries),
the relative frequency of punctuation and space characters per element (to identify, e.g., JSON or XML elements, which are structured by those characters),
the relative frequency of elements (categorical variables would have a low proportion of unique values in comparison to other variables).
util_string_is_not_categorical(vec)
TRUE or FALSE
a character vector