Fix regressions from the iD editor upgrade and cross-editor bleed fix (#7279)

[sumitdahal7]

What type of PR is this? (check all applicable)

• 🍕 Feature
• 🐛 Bug Fix
• 📝 Documentation
• 🧑‍💻 Refactor
• ✅ Test
• 🤖 Build or CI
• ❓ Other (please specify)

Describe this PR

After #7279 (the @openstreetmap/id 2.38.0 → 2.41.2 upgrade + the fix for OSM iD / Sandbox editors bleeding into each other), new issues showed up, and this PR addresses one without reverting the original fix.

1. Recently-saved edits not showing without a hard reload
#7279 fixed the bleed by nulling the context on every unmount, forcing a brand new one on every reopen but that also meant a task you just edited and saved would come back from a cold, empty cache instead of the warm one that used to show your own recent changes. On top of that, a fresh context's initial tile fetch could get aborted by our own cache reset, with nothing left to re-request it.

• Contexts are now tagged by which editor built them and only replaced when the type actually changes, so same-editor reopens stay warm; a fresh context's view is explicitly redrawn after its connection is set up, so the task's own data reliably gets re-fetched instead of staying blank.

2. Replace Risky Stylesheet toggling with tagged iD contexts
The stylesheet-disable mechanism from the previous commit could tag the wrong in a real session and bleed the whole app's CSS.

• Replaced with tagging each context by which editor built it, reused only when the type matches.

Updates
Shared logic is extracted into utils/idEditorContext.js (also resolves theSonarCloud duplication).

[sonarqubecloud]

Quality Gate passed

Issues
2 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud