also add byte indices to diagnostic ranges

[WebFreak001]

For tools wanting to read from the source file this makes it much easier to look up the code.

Before we tag the new API with start and end locations as new tag, we better just add this together with everything, since then we don't need to change the API again later. Grouping together lines / columns as size_t[2] also makes the previous API better to understand, e.g. because of size_t line, size_t startColumn, size_t endColumn now being size_t line, size_t[2] columns