Remove unused track_idX columns from CSV table export after merger resolving

[chaubold]

Before we had proper merger resolving (or when it is switched off), we added several columns to the exported CSV table to list all the track_ids contained in a single merger (and inconsistently did not store several lineage ids). Now that merger resolving is working nicely, we add new rows for the newly created resolved objects. But the columns for having multiple track_ids are still there. These could be removed.