Add lightweight reviewer app and review endpoints

[chezhia]

Summary

Adds a lightweight MONAILabel reviewer workflow for existing segmentations without loading AI tasks.

What changed

• adds a review-focused sample app under sample-apps/reviewer with the required MONAILabel app entrypoint
• adds reviewer aggregate endpoints under /review and mounts them in the main FastAPI app
• aligns the reviewer client and smoke test with the implemented server routes
• fixes reviewer image downloads in the 3D Slicer plugin by using direct REST fetches and normalizing server URLs at the UI and REST layers
• teaches local datastore reconciliation to recognize reviewer label files named .seg.nrrd
• documents the reviewer app in the top-level README and refreshes the sample reviewer README to match the implemented workflow

Validation

• python -m unittest tests.unit.datastore.test_local
• python -m py_compile monailabel/app.py monailabel/datastore/local.py monailabel/endpoints/datastore_review.py sample-apps/reviewer/app.py sample-apps/reviewer/client.py sample-apps/reviewer/main.py sample-apps/reviewer/lib/config.py plugins/slicer/MONAILabelReviewer/MONAILabelReviewer.py plugins/slicer/MONAILabelReviewer/MONAILabelReviewerLib/ImageDataController.py plugins/slicer/MONAILabelReviewer/MONAILabelReviewerLib/MonaiServerREST.py test_reviewer_server.py

Notes

• untracked scratch notes and draft cache-helper files were intentionally left out of this PR because they are not wired into the current reviewer workflow.

Summary by CodeRabbit

• New Features
  • Added authenticated /review APIs for listing review cases, viewing per-image label versions, and generating review reports (json|csv|html) with filtering.
  • Introduced a lightweight Reviewer sample app and client for review/approval workflow and report generation.
• Bug Fixes
  • Improved .seg.nrrd label-to-image ID matching and standardized review timestamp (last_reviewed) behavior.
  • Hardened reviewer plugin server URL handling, DICOM loading, and temporary file path safety.
• Documentation
  • Expanded README with Reviewer App onboarding and added dedicated reviewer sample app documentation.
• Tests / Quality
  • Added unit tests for review endpoints, reviewer app/client behavior, datastore logic, and date-range handling, plus a manual smoke test.
• Chores
  • Updated tooling and pinned dependencies for highdicom by Python version.