The goal of this issue is to make date range archiving faster, by letting the archiver process on-demand only the requested data set when idSubtable is used.

Notes:

• currently idSubtable is an integer. We would need to change it to be a string representing the label and path to this subtable
  • this is needed because we can generate the subtableIds only if we build the expanded table, but here we want to avoid processing this expanded table.
• given an idSubtable, we could only fetch the requested recordName and only the requested 1st level table and the N tables needed for this sub-table.

(It's not an easy one unfortunately.)

Follows up #4768