check: optimization potential with --verify-data AND --undelete-archives

[ThomasWaldmann]

check --verify-data does a full scan over the repository, just reading and authenticating all objects (but not yet using any of the data it read).

it could check if the chunk is an archive metadata chunk and if so, remember its ID.

undelete-archives would normally do a full scan over the repo also, trying to identify archive metadata chunks - it could use the remembered list from verify_data (if that option was given).

[ThomasWaldmann]

duplicate of #8478.