♿ Restore focus rings and wire tab ARIA relationships

[chrisjsewell]

Carries #230 (thanks @gabalafou!) forward onto the current Sass layout, and adds programmatic label→panel references for tabs.

Changes

• Focus rings restored (from Restore focus rings #230, ported hunk-by-hunk onto the post-@use partials, all six files): :focus-visible rings on dropdown summaries, tab labels, buttons and stretched-link cards; 0.25rem ring offset; ring colors matching element colors; the tap-highlight typo fix (:not(.focus-visible) class → :not(:focus-visible) pseudo); tab-label top-padding evened out. Rings appear for keyboard focus only — mouse clicks show none.
• Tab ARIA wiring (Include radio/label ID as part of tab-content div #30): each tab content panel gets a stable id (sd-tab-item-N-content, prepended so anchor ids from preceding hyperlink targets keep working), and each label emits aria-controls pointing at it. This establishes the relationship in the accessibility tree; it is deliberately not a full WAI-ARIA tabs pattern (the CSS-only radio mechanism is kept, and AT support for aria-controls varies).

Verification

• Independent review verified the port faithful for all 11 source hunks, zero compiled-CSS drift against the SCSS, and that the HTML surface diff vs main is exactly the intended set. The multiple-ids case was force-tested through the docutils HTML5 writer: the -content id always wins the id attribute, so aria-controls always resolves and old anchors keep working.
• Playwright keyboard smoke on built docs: Tab-focus produces a visible ring (:focus-visible true, outline-style: auto); mouse click produces none; 24/24 labels' aria-controls resolve to real panels. (axe-cli was unavailable in the sandbox — its chromedriver download is blocked; manual steps documented in the test file.)
• Full suite on pinned Sphinx 7.2.6 / 7.4.7 / 8.2.3 (173 passed each); new tests/test_tabs_a11y.py; 5 regression fixtures regenerated with only the expected id/aria attribute changes.

Closes #30
Closes #230 (carried, with author credit)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.84%. Comparing base (3adada6) to head (dedf382).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #283      +/-   ##
==========================================
+ Coverage   90.81%   90.84%   +0.03%     
==========================================
  Files          13       13              
  Lines        1219     1224       +5     
==========================================
+ Hits         1107     1112       +5     
  Misses        112      112              

Flag	Coverage Δ
pytests	90.84% <100.00%> (+0.03%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.