Require Algorithms & CategoryMaps to be registered before processing

[mihow]

Summary

Formerly we allowed prediction responses to define the algorithm & category map they used. This was flexible but added a lot of complexity. Now the algorithms and category maps must defined ahead of time and a PipelineNotConfigured exception will be thrown if an unexpected algorithm is returned from the processing service. The pipeline registration already handles the checking & creation of the algorithms available from a processing service.

Related Issues

Relates to #782 & #721. Closes #941

Deployment Notes

• Remove the AlgorithmCategoryMap from each prediction response in the live processing service (the ADC). This should improve performance as well by greatly reducing the size of the serialized object returned & passed to the save_results task. Currently every detection in every image includes a list of all classes in each response.
  The changes on the ADC side are now done and deployed Remove algorithm and category map details in every result response ami-data-companion#93

Checklist

• I have tested these changes appropriately - tested on live db snapshot with latest ADC.
• I have added and/or modified relevant tests.
• I updated relevant documentation or comments.
• I have verified that this PR follows the project's coding standards.
• Any dependent changes have already been merged to main.

[netlify]

✅ Deploy Preview for antenna-preview ready!

Name	Link
🔨 Latest commit	c7516d8
🔍 Latest deploy log	https://app.netlify.com/projects/antenna-preview/deploys/68cb596b3005510008848a8b
😎 Deploy Preview	https://deploy-preview-942--antenna-preview.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
Lighthouse	1 paths audited Performance: 64 (🟢 up 4 from production) Accessibility: 89 (🟢 up 9 from production) Best Practices: 92 (🔴 down 8 from production) SEO: 100 (🟢 up 8 from production) PWA: 80 (no change from production) View the detailed breakdown and full score reports

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[Copilot]

Pull Request Overview

This PR enforces a stricter pipeline configuration by requiring algorithms and category maps to be pre-registered before processing, removing the previous flexibility of dynamically creating them from pipeline responses. This change simplifies the system by throwing PipelineNotConfigured exceptions when unregistered algorithms are encountered.

• Algorithms and category maps must be defined ahead of time via pipeline registration
• Unknown algorithms returned from processing services now raise exceptions instead of being dynamically created
• Processing service dependency added to docker-compose to ensure proper initialization order

Reviewed Changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
docker-compose.yml	Adds ml_backend dependency to ensure processing service is available before pipeline registration
ami/ml/exceptions.py	Introduces new PipelineNotConfigured exception for better error handling
ami/ml/schemas.py	Renames PipelineConfig to PipelineConfigResponse for clarity
ami/ml/models/algorithm.py	Adds task type enums, category map validation, and improved manager
ami/ml/models/processing_service.py	Updates to support pipeline configuration parameter passing
ami/ml/models/pipeline.py	Major refactoring to use pre-registered algorithms instead of dynamic creation
ami/ml/tests.py	Updates tests to verify new exception behavior and removes dynamic algorithm creation tests

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}