DataFrame and DataSeries should not perform type checks

[olekscode]

The current version of DataFrame knows the type of data stored in each of its columns at all times. Every time something is changed (a value is added, removed, or updated), DataFrame checks all types again to keep them up to date. This takes a lot of time.

Same happens with DataSeries.

I think that type checks should not be done in Pharo. Instead of checking if a column contains numbers, we should expect it to behave as if it contains numbers and throw exception otherwise.

Therefore, DataFrame>>columnTypes, DataSeries>>type, and all the related functionality should be removed.