test: gate HTML tree structure a browser would silently repair

[hivecommons-hive]

Test Improvement

Adds scripts/page-structure.test.mjs, a zero-dependency tokenizer/tree gate over every committed HTML page (redirect shortcuts and 404.html included). The existing gates tokenize tags one at a time with /<tag[^>]*>/ and the stub-DOM tests build their own elements, so nothing noticed markup a browser repairs silently. The gate parses each page the way the HTML tokenizer does — comments, raw-text <script>/<style>/<textarea>/<title> bodies, quoted attribute values, the spec's optional end tags (<li>, <dt>/<dd>, <p>, <option>, table parts), foreign content for inline SVG — and reports every repair a browser would make:

• stray, mismatched or missing end tags (with both line numbers), </br>, self-closing <div/> outside SVG, unclosed comment/<script>;
• a block element inside <p>, reported as the stray </p> it leaves behind;
• the same attribute twice on one element (second is dropped);
• a raw </> inside an attribute value — valid HTML, but it truncates the tag that every sibling [^>]* gate reads, so this also hardens those;
• <li>/<option>/<tr>/<td>/<figcaption>/<summary>/<legend>/<source>/… outside their required parent;
• interactive content nested in <a>/<button>/<label>;
• non-metadata content in <head>;
• exactly one <html>/<head>/<body>, at most one <main> (required on non-redirect pages).

Each rule has a fixture self-test. Picked up by the existing node --test scripts/*.test.mjs step — no workflow change. README ## Checks documents the gate (required by ci-wiring.test.mjs).

Verified locally at cb88619: full suite 351 → 476 pass; check-links.sh --external-warn, check-links.test.sh, check-redirects.test.sh, check-redirects.sh all green. Mutation checks against index.html (each restored afterwards): deleting one </section> → line 646: </main> but <section> opened at line 60 is still open; a second class= on an <a> → repeats attribute "class"; > in the description content → contains a raw "<" or ">"; <button> inside <a> → nested inside <a>. Every pre-existing gate stays green on all four.

Closes #138

---

Filed by quality agent (ACMM L4/L6 — full mode)

— hive: agent=quality backend=copilot model=claude-fable-5.1 copilot=1.0.88

[kubestellar-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment