use shapeParameters in MVA calibration

[giovannimarchiori]

This PR adds to the MVA calibration algorithm the possibility to read the energy fractions per layer directly from the shapeParameters of the input clusters (if available) rather than recalculating them on the fly. It also shows how to save in output the shapeParameters of the input collections plus some extra parameters (in this case the raw energy of the cluster before the MVA calibration).

[giovannimarchiori]

Usage example: very similar to before, just need to pass to the algorithm the string of the detector name used in the shapeParameters to label the energy fraction decorations, i.e. add the systemNames = ... row below:

calibrateCaloClusters = CalibrateCaloClusters("calibrateCaloClusters",
                                              inClusters="Augmented" + createClusters.clusters.Path,
                                              outClusters="Calibrated" + createClusters.clusters.Path,
                                              systemIDs=[4],
                                              systemNames=["EMB"],
                                              numLayers=[ecalBarrelLayers],
                                              firstLayerIDs=[0],
                                              readoutNames=[
                                                  ecalBarrelReadoutName],
                                              layerFieldNames=["layer"],
                                              calibrationFile="data/lgbm_calibration-CaloClusters-v03.onnx",
                                              OutputLevel=INFO
                                              )