fix(html): prepend floating color-scheme toggle

[johnfinnerty-nz]

Description

Place the floating color-scheme toggle container first in the body so its DOM order matches its visual position. Add a Playwright assertion that protects the insertion order.

Closes #14838

Testing

• npx playwright test tests/html-color-scheme-toggle.spec.ts --list (9 tests discovered)
• npx prettier --check src/resources/formats/html/templates/quarto-html-after-body.ejs tests/integration/playwright/tests/html-color-scheme-toggle.spec.ts
• Source contract check and git diff --check

Full browser execution was not run because the generated fixture sites and browser dependencies are not available in this checkout.

Checklist

I have (if applicable):

• referenced the GitHub issue this PR closes
• updated the appropriate changelog in the PR
• ensured the present test suite passes
• added new tests
• created a separate documentation PR in Quarto's website repo and linked it to this PR

AI-assisted PR

• AI tool used: OpenAI Codex
• Codebase grounding: Local clone of quarto-dev/quarto-cli
• Human review: John Finnerty reviewed the prepared change summary and authorized submission. Verification performed by Codex is listed above.

[posit-snyk-bot]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scan Engine	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues
✅	Licenses	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.

[cscheid]

What bug does this fix? We're not inclined to make changes to everyone's web pages with little reason.

[johnfinnerty-nz]

Thanks for asking. The issue I was trying to address is the focus-order mismatch documented in #14838: on a plain document, the floating color-scheme toggle is drawn at the top right but appended to the end of body, so keyboard and screen-reader users reach it after the page content. It is a usability concern, not a WCAG failure.

I agree that changing DOM order on every affected page needs stronger regression evidence, especially around theme selectors and full browser behavior. I did not complete that validation, so I am closing this rather than asking the project to accept an under-tested global change. Sorry for the noise.