Update ML training API reference for training container methods

[Shannon Bradshaw (shannonbradshaw)]

Source changes

• APP-18127, APP-17611 — viam-python-sdk#1267 (08502a67): MLTrainingClient gains list_containers(org_id), get_container(id), register_custom_training_container(org_id, image_uri, description) and delete_custom_training_container(id), and submit_custom_training_job gains an optional container_id. Released in Python SDK v0.83.1 (git tag --contains 08502a67), and now on python.viam.dev.
• Server side: app#13601, app#13637, app#13784 (GetContainer, APP-18128) and app#13641 (container_id on SubmitCustomTrainingJob, APP-17651) are live.

Documents: DOCS-5002 (APP-18128, GetContainer). Also matched: the epic-level DOCS-4913 (Blocked) and DOCS-5010 (APP-17651).

Docs changes

• .github/workflows/sdk_protos_map.csv: new mltraining rows for ListContainers, GetContainer, RegisterCustomTrainingContainer and DeleteCustomTrainingContainer, Python column only. The Go MLTrainingClient in RDK v1.11.0 has none of these methods, TypeScript v0.76.0 has none, and the generator has no Flutter app-client support.
• static/include/app/apis/overrides/protos/mltraining.*Container*.md: one-sentence descriptions for the four methods. The register and list descriptions carry the server rules from app mltraining/ml_training_service.go (publicly readable linux/amd64 Docker Hub images only; Viam's public containers plus the organization's own).
• static/include/app/apis/generated/mltraining.md, mltraining-table.md: regenerated. The diff adds the four methods and the container_id parameter on the Python submit_custom_training_job tab, and nothing else.

Generator note: I ran update_sdk_methods.py "" app,mltraining rather than "" mltraining. With mltraining alone, parse_go.py only scrapes Go app-client methods when the app resource is in the run, so every Go tab on the page disappeared. Adding app scrapes them; app.md came out unchanged, so this PR touches only the mltraining files.

Known gap, not changed here: the scraped Python docstring marks description as optional on register_custom_training_container, but the server rejects an empty description (errNoDescriptionSpecified). That text comes from python.viam.dev, so the fix belongs in the SDK docstring.

How I found these

• Xref lookup: app-api-xref.md (MLTrainingService) and the docs-agent backlog item python-sdk-08502a67-ml-training-containers, whose two release conditions both hold as of this run.
• Grep matches: no other page documents these SDK methods.

Checks: make build-prod passes. Generated files are outside prettier and markdownlint scope.

---

Generated by daily docs change agent

🤖 Generated with Claude Code

https://claude.ai/code/session_01J5WsNbGP1yKuqz7GyobaFN

---

Generated by Claude Code

[netlify]

✅ Deploy Preview for viam-docs ready!

Name	Link
🔨 Latest commit	c8225a9
🔍 Latest deploy log	https://app.netlify.com/projects/viam-docs/deploys/6ac6124a0549620008bf326f
😎 Deploy Preview	https://deploy-preview-5398--viam-docs.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
Lighthouse	1 paths audited Performance: 44 (🟢 up 1 from production) Accessibility: 100 (no change from production) Best Practices: 100 (no change from production) SEO: 92 (no change from production) PWA: 60 (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.

[Jeremy Rose (jeremyrose-viam)]

Follow-up on the known gap in the description: the description parameter of register_custom_training_container shows as optional because the Python SDK declares description: str = "", but the server rejects an empty value. Upstream fix proposed: viamrobotics/viam-python-sdk#1289. Once it ships in a Python SDK release, regenerating mltraining.md will mark it required.