* Add CoreML instance segmentation configs * Add common configuration for Core ML backend * Fix pre-commit hook failures