docs: stop publishing the internal REVIEW.md code-review artifact

[sorenwacker]

docs/REVIEW.md is a point-in-time internal review artifact, but it is currently published to the public docs site at https://sorenwacker.github.io/metaseed/REVIEW/ (HTTP 200).

Two problems with shipping it:

1. It is stale and misleading. It presents two findings as confirmed-open HIGHs that are already fixed on main:
   • cli/app.py:251 "convert resolves --profile but ignores it" — fixed; the call now reads get_model(entity, version, profile).
   • repositories/memory.py:167 "update_entity returns stale data" — fixed; it now returns self._node_to_entity(updated or node, ...).
2. It leaks a local filesystem path — /Users/sdrwacker/workspace/metaseed-project/... is visible on the live page.

Why exclude_docs and not just a nav removal

MkDocs renders every page under docs/, whether or not it appears in nav. Dropping the nav entry alone would keep the page live at the same URL. exclude_docs (MkDocs ≥1.6; the project pins mkdocs>=1.6,<2.0) removes it from the build itself.

The file stays in the repo — only the published site changes. This deliberately avoids a conflict with the open #97, which edits docs/REVIEW.md.

Verification

Built the site both ways with mkdocs build --strict, exactly as .github/workflows/docs.yml does:

	site/REVIEW/index.html
before	published
after	gone

Strict build passes and the other 48 pages still build.

Noted, not fixed here

docs/isa_demo.ipynb has the same class of leak — it is not in the nav yet is live at https://sorenwacker.github.io/metaseed/isa_demo.ipynb, exposing /Users/sdrwacker/workspace/miappe-tools/metaseed/src/metaseed/specs/isa_v1.0.yaml. That path also references a defunct project name and the pre-<profile>/<version>/profile.yaml spec layout, so the notebook may be stale. Left alone pending a decision on whether to regenerate or drop it.

🤖 Generated with Claude Code

https://claude.ai/code/session_01FTZQT3bQzqTBXQXvDsk4w9