id offset for generator functions

[henrymartin1]

The trackintel generator functions (generate_triplegs, generate_trips, ...) assign an ID to every instance. This ID plays a key role in many trackintel functions. If we want to merge trips from different sources/datasets to processes them jointly, we have to make sure that all trips have unique IDs. Therefore it would be great to have an ID offset as a parameter that allows IDs that do not start at 0.

[bifbof]

I am kind of against it, I see the motivation but adding the extra keyword is in my opinion not much more convenient than just using gdf.index += offset.