fix(data-grid): wait for nested shadow components before measuring

[amir-ba]

Summary

Fix the remaining data-grid readiness issue from #2552. Wait for each component before traversing its light-DOM and shadow-DOM children, so children created during hydration are included before auto-width measurement.

 calculateAutoWidths
-  await light-DOM descendants
+  await each cell recursively through light DOM and shadow roots
   await fonts
   measure widths

Extend the existing regression test with a delayed component inside two shadow roots. Keep the existing light-DOM case and stale fields/rows guard. No public API change.

Evidence

• Before: The new nested-shadow case fails: expected width auto, received 12 before hydration completes.
• After: Both readiness cases pass and the hydrated width is 120.
• Core build and lint pass.
• Core specs: 62 suites, 360 tests, 168 unchanged snapshots.
• Core E2E: 35 suites, 78 tests pass.
• Pinned Linux Chromium data-grid visuals: 26 cases pass in light/dark mode with snapshot updates disabled. No baselines changed.
• Changed-file Prettier and whitespace checks pass.

Merge Danger

Door: two-way

Blast Radius: data-grid

Only automatic column measurement timing changes. Existing visual baselines are unchanged. A descendant that never resolves componentOnReady() will continue to delay measurement, as a light-DOM descendant already does.

[Copilot]

🟢 Approval recommended

The implementation matches the stated scope and provides focused regression coverage without introducing unresolved issues.

0 open findings

What changed in this PR

Ensures DataGrid auto-width measurement includes content hydrated within nested shadow roots.

Changes:

• Recursively waits for light- and shadow-DOM descendants before measurement.
• Extends regression coverage for light DOM and two nested shadow roots.

File	Description
data-grid.tsx	Adds recursive component readiness traversal.
data-grid.spec.ts	Tests delayed hydration across light and nested shadow DOM.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.