Prep for new Models

[robert-graf]

No description provided.

[Copilot]

Pull request overview

This PR prepares the segmentation pipeline for new model variants by extending inference configuration (C1/sacrum handling) and adding optional auto-cropping/timing instrumentation around the semantic/instance pipeline.

Changes:

• Extend Segmentation_Inference_Config with has_c1 and sacrum_ids, and thread these into postprocessing.
• Add optional VIBESeg-based auto-cropping and per-stage timing logs to process_img_nii.
• Refactor/robustify parts of semantic/instance/postprocessing (e.g., canal-empty guard, CC assignment logic, label mapping cleanup).

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
spineps/utils/seg_modelconfig.py	Adds new inference-config knobs (has_c1, sacrum_ids) to support upcoming model label sets.
spineps/seg_run.py	Introduces auto-crop + timing, CT-specific preprocessing tweak, and passes new config values into postprocessing.
spineps/seg_model.py	Minor cleanup and improved failure logging in model run().
spineps/phase_semantic.py	Adds a guard for empty canal segmentation in inferior-cleaning logic.
spineps/phase_pre.py	Adds compute_crop() wrapper around VIBESeg with backward-compat handling.
spineps/phase_post.py	Adds disable_c1/sacrum_ids, refactors postprocessing order and missing-CC assignment logic.
spineps/phase_labeling.py	Removes unused sacrum-related snippet in labeling step.
spineps/phase_instance.py	Adjusts corpus detection checks, improves logging, and simplifies label mapping/cutout code.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.