Merge pull request #279 in PIPE/pipeline from PIPE_673_VLA_heuristics_fixes to master
* commit '4d8f8797eaf3e044154f05be03a5b32c2b9839bf':
fix: PIPE-674: pblimit_cleanmask == pblimit_image leads to empty arrays when computing stats for weblog. Revert pblimit_cleanmask to base class value.
Merge pull request #232 in PIPE/pipeline from PIPE-682_VLA_niter_heuristic_update to master
* commit '835fe35981e7e84e8f28068c41eeef9b2d78a6b1': PIPE-682: remove beam parameter in niter_correction child class definitions. PIPE-682: defer passing actual synthesized beam size to ninter_correction() method to the next development cycle PIPE-682: Merge any newly computed beams with the context. PIPE-675: VLA niter heuristics
chg: mask_frac_rad=0.45 for both ALMA and VLA chg: ensure that estimated niter is smaller than 2**32 PIPE-682: VLA niter heuristic update: VLA should use similar heuristics as already implemented for ALMA with assuming a different mask radius.
Merge pull request #271 in PIPE/pipeline from PIPE-681_VLA_uvrange_heuristic_update to master
* commit 'dc25be416e48706b8f675be78ab94bc29445bf23': PIPE-681: VLA uvrange heuristics update PIPE-681: VLA uvrange heuristic update: when the amplitude in smallest baseline bin (the 5% of shortest baselines) is more than 2x the amplitude of the 50-55% bin, uvrange excludes the shortest 5% of baselines.
PIPE-681: VLA uvrange heuristics update
chg: assume that the largest baseline is associated with the highest frequency spw
chg: use real spwids instead of virtual spwids when obtaining the amplitude and baseline statistics
