[2026 AI] Move model weights out of src and version them with provenance metadata

[Nanle-code]

Objective

Treat trained models as versioned artifacts with known provenance instead of loose files in the source tree.

Target area

src/lib/model_weights.json, src/ml

Context

src/lib/model_weights.json sits next to application code with no record of the training data, code version or date that produced it, and changes to it are nearly impossible to review.

Acceptance criteria

• Weights move to models/<name>/<version>/ (or release assets) with a model.json manifest: training commit, dataset hash, metrics, created date.
• The app loads models by manifest version, and a checksum is verified at load time.
• The training script writes the manifest automatically.
• PR review guidance for model updates is documented.
• Automated tests cover the primary flow, at least one boundary case, and at least one failure case.
• User-facing documentation or developer guidance is updated, including compatibility, security, or migration notes where applicable.

Pull request requirements

• All continuous integration checks must pass before a pull request may be merged.
• The branch must be free of merge conflicts with the target branch at the time of merge.
• Do not request merge with failing workflows, skipped required checks, or unresolved conflicts.

[drips-wave]

@Nanle-code This issue could not be added to the Stellar Wave Program because the per-repo points budget would be exceeded (25000 / 25000 points already used). The label has been automatically removed.

The budget resets at the end of each Wave. You can review the remaining budget for your repos on the Orgs & Repos page.

ℹ️ Learn more about points budgets