Revert "Use piff for characterizeImage in DRP"

Now that we're using the final Piff PSF measurements to drive coadd
selection and other visit-summary-based things, we can switch back to
PSFEx for speed and robustness for our first PSF model.

ingredients/DRP-minimal-calibration.yaml

@@ -7,12 +7,7 @@ description: |
 
 tasks:
   isr: lsst.ip.isr.IsrTask
-  characterizeImage:
-    class: lsst.pipe.tasks.characterizeImage.CharacterizeImageTask
-    config:
-      - python: |
-          import lsst.meas.extensions.piff.piffPsfDeterminer
-          config.measurePsf.psfDeterminer = "piff"
+  characterizeImage: lsst.pipe.tasks.characterizeImage.CharacterizeImageTask
   calibrate: lsst.pipe.tasks.calibrate.CalibrateTask
   writeSourceTable: lsst.pipe.tasks.postprocess.WriteSourceTableTask
   transformSourceTable: lsst.pipe.tasks.postprocess.TransformSourceTableTask