Allow duplicate headers in CSV File

Currently if a CSV file has a duplicate column header, a runtime exception is thrown:

https://github.com/apache/commons-csv/blob/546fd69eaa6b3fd2c9da8b4222a3822f1ab53abd/src/main/java/org/apache/commons/csv/CSVParser.java#L492

I think that may be a bit harsh reaction, as there could be some columns whose data is relative to its column order. I understand the header map uses the column name as the key, however I anticipate one could flip the map to be a Map<Integer, String> so its column number is the key.

Anyway I am curious what the genesis of this decision was to use column headers as the key and if it's worthwhile to flip the map and allow for duplicate column names.