Fix launching work directly (not via Slurm)

Description

The --no-cluster-job option was failing to even start jobs. Closes #58.

How Has This Been Tested?

Launched on max-exfl-display:

xfel-calibrate AGIPD CORRECT \
    --in-folder /gpfs/exfel/exp/SPB/202131/p900230/raw --run 174 \
    --out-folder /gpfs/exfel/data/scratch/kluyvert/agipd-calib-900230-174-direct \
    --karabo-id SPB_DET_AGIPD1M-1 --karabo-id-control SPB_IRU_AGIPD1M1 --karabo-da-control AGIPD1MCTRL00 \
    --modules 0 --sequences 0 --no-cluster-job

I let this run until it saved the first HDF5 output files, and then cancelled it (Ctrl-C) so as not to hog the login node. Before these changes, it was failing to even start running the notebooks.

Types of changes

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• My code follows the code style of this project.

Reviewers

@ahmedk @danilevc

src/xfel_calibrate/finalize.py

@@ -141,6 +141,8 @@ def get_job_info(jobs: List[str], fmt: List[str]) -> List[List[str]]:
     Result ordered according to order of jobs given
     Order of fields in inner lists follows fmt
     """
+    if not jobs:
+        return []  # Skip calling sacct if not using Slurm
 
     # will use JobID to match results to jobs (duplicate field in fmt is OK)
     fmt_query = ",".join(["JobID"] + fmt)
@@ -389,17 +391,6 @@ def tex_escape(text):
 def finalize(joblist, finaljob, run_path, out_path, version, report_to, data_path='Unknown',
              request_time='', submission_time=''):
     run_path = Path(run_path)
-    print("Waiting on jobs to finish: {}".format(joblist))
-    while True:
-        found_jobs = set()
-        output = check_output(['squeue']).decode('utf8')
-        for line in output.split("\n"):
-            for job in joblist:
-                if str(job) in line:
-                    found_jobs.add(job)
-        if not found_jobs:
-            break
-        sleep(10)
 
     prepare_plots(run_path)