[test][avatar] Add axe tests and docs

[mj12albert]

Result	Count
✅ Supports	9
⚠️ Partially Supports	2
❌ Does Not Support	0
➖ Not Applicable	44

Doc: https://github.com/mj12albert/material-ui/blob/avatar-axe-coverage/packages/mui-material/src/Avatar/accessibility.md

• I have followed (at least) the PR section of the contributing guide.

[code-infra-dashboard]

Deploy preview

https://deploy-preview-48732--material-ui.netlify.app/

Bundle size

Bundle	Parsed size	Gzip size
@mui/material	0B(0.00%)	0B(0.00%)
@mui/lab	0B(0.00%)	0B(0.00%)
@mui/private-theming	0B(0.00%)	0B(0.00%)
@mui/system	0B(0.00%)	0B(0.00%)
@mui/utils	0B(0.00%)	0B(0.00%)

Details of bundle changes

---

Check out the code infra dashboard for more information about this PR.

[michelengelen]

one single finding from claude.
I reviewed the .md file and it looks clean.

Can be merged as ist or with a tighter guard around the test, your call!

[michelengelen]

Array.from(...).find(node => node.textContent === 'OP') returns undefined if the LetterAvatars fixture is renamed, the "OP" avatar text is changed, or the route resolves incorrectly. The immediately following range.selectNodeContents(avatar) throws a TypeError (DOM spec requires a Node). Because style.remove() sits after the throw point, the injected * { line-height: 1.5 !important; letter-spacing: 0.12em !important; word-spacing: 0.16em !important; } style sheet persists in document.head for all subsequent Playwright tests in the same browser session, silently corrupting every screenshot taken afterward. Fix: wrap the measurement block in try { … } finally { style.remove(); } and add an explicit assertion that avatar is non-null before proceeding.

[michelengelen]

Closing in favour of #48917.

@mj12albert has left the company, so this branch can no longer be rebased or
updated. To keep the work alive, the series has been rebased onto current master
and reopened from a maintained fork. Commit authorship is preserved — the
component commits are still authored to Albert.

The full series is #48915 → #48926, ending in a public accessibility conformance
report (#48926) that answers #14187. Nothing from this PR is lost; the report,
tests, and demos all carry over.

Thanks @mj12albert for the original work.