updating MPI sim layer for blocking receives, updating dragonfly model for the message remainders, checking for negative byte size in dumpi traces

so people can bring their own workload

currently ignored by all

- fix noop workload ops
- get rid of unused workload option

note: delays in dumpi are now in nanoseconds rather than seconds

Note that the recorder workload now requires a parameter for the
number of ranks in the original application. the workload dump
utility has been updated to reflect this.

Darshan and recorder workloads have temporary debug parameters that
are no longer used and can be removed.

Workload API now defines a function for retrieving the rank count
of a workload. Currently, only darshan workload implements this.
Will work on bgp workload and recorder later.

Pull all darshan workload generator logic into the workload generator
API. Nondeterminism (i.e. using rand()) still needs to be removed from
generating collective I/O events, since they will now be generated in
a collective manner -- not independently as before. Also, still need
to sort out collective timing issues.

Modifications to autotools to include necessary info for building
the darshan workload API. This API compiles fine, but the mpi replay
test is not linking properly with darshan library.

updates to codes workloads: switched to hash tables in the workload--i/o lang plug-in, removed codes_workload_get_info from i/o lang and darshan workloads as this is no longer needed in the bg/p model

Updating the workload test to use the config file for the 'bgp_io_workload' and 'test' workload types. Loads the workload file for the bgp_io_workload type. Tests for remaining operations are to be added