Refactor AbstractRow, Row, and HeaderRow

[nozavroni]

UPDATE for v0.3:

Version 0.3 is going to be a major overhaul of the way data is handled in CSVelte. Basically rather than try to create a table inside of the reader/writer, you will create a dataset (table) and then populate it with a reader and write it to disk with a writer. (See issue #46)

So, in that context, this issue will require that I remove Table\Row and Table\HeaderRow completely from CSVelte\Reader and CSVelte\Writer. They will be moved into CSVelte\Table\Dataset or something similar.

• [ ] Remove Row and HeaderRow from Reader
• [ ] Remove Row and HeaderRow from Writer
• [ ] Figure out how this separation a little more... how can you separate the two without having to duplicate code between them.
• [ ] Finish this list later

---

They are causing a bunch of problems with the writer and I think the majority of it stems from too much logic happening in AbstractRow when it really belongs in either HeaderRow or Row.

[nozavroni]

Reopening because of issues mentioned in updated issue description.