Enhance the MOVE DATA and RECLAIM STG commands

A version 6 server doing dedup presents new
challenges to the admin to manage the
server's storagepools. These enhancements
will enable the admin to target whole volume
operations to achieve maximum thruput.

Improve the volume selection for both input
and output by adding CRITERIA parameters to
RECLAIM STG and MOVE DATA commends. (See the
previous requirement for CRITERIA objects).
This will allow a single filesystem to be
targeted for input and output. The result -
when commands are scheduled properly - will
be a single operation per raid array.
CRITERIA parameters will select multiple
volumes thus reducing the admins workload.

In addition, for RECLAIM STG allow multiple
commands to be issued on the same storagepool
when CRITERIA are in use.

In addition, for MOVE DATA, implement a fast
copy feature where the OS cp comand (or
similar function added to the server) is
invoked to move a volume from one filesystem
to another, with no analysis of the volumes
contents, to rapidly empty the filesystem
when the hardware is at end-of-life.

IMPLEMENTATION SUGGESTION.

reclaim stg <pool-name> inputvols=<criteria-
spec> outputvols=<criteria-spec> [dur=nn
th=nn wait={y/n} ]

move data inputvols=<criteria-spec>
outputvols=<criteria-spec> oscopy={n/y}

(NOTE: I only do reclaims with the command, I
am not interested having this fitted into
automatic reclaim.)