This very simple tool dumps all samples in a LevelDB metrics storage
together with their chunk information. This is very crude for now - I
just whipped it up because I needed to debug something. We should add
printf-style formatting support for it in the future to indicate what
information to output for each sample in the database. We could also
support inspecting the index databases.
This commit simplifies the way that compactions across a database's
keyspace occur due to reading the LevelDB internals. Secondarily it
introduces the database size estimation mechanisms.