Parallel archive access

The performance when accessing large archive query result sets is poor. We identified this is mostly due to many small serial file system accesses to GPFS. The endpoint nomad.app.api.archive.ArchiveQueryResource should use threading to read for many calcs simultaneously.