Allow greater control of data representation #8
Description
Example adapted from PR #7:
Data: "Sydney is a city in Australia."
When labelling "Sydney" as a "LOC", we currently get:
(0, 0) - LOC
It would be nice to have the ability to specify that we always have a start and end to spans (for easier processing elsewhere).
Similarly, it would be nice to be able to specify that the character position should be included, even when annotating tokens.