fix so this makes it possible to actual run this test if you have more than 345 processes running, which could easily happen from my experience.

- the select_job_res structure has been renamed job_resources
- all pertinent references to "select_job_res" have been renamed "job_resources"
- the select_job_res_t *select_job member of the slurmctld.h's job_record has
    been renamed job_resources_t *job_resrcs

When querying database for jobs in certain state and a time period is given only jobs in that state during the period will be returned, previously if a time period was given in sacct jobs eligible to run or running would be displayed, which is still the default if no states are requested.

changed code back to original so regression whould work.  It appears job_min_cpus really means min cpus per node not job min cpus.

added logic for modify job to work correctly on bluegene systems, also added check to make sure limits are abused on a job update

    name of NULL rather than returning no information about the node so that
    node index information is still valid.