fix: resolve retry/resume issues + code cleanup

[sharanyavinod]

• Fixes error scenario when socket is dropped during questions/user inputs
• Muted ux on schema retry errors
• Code cleanup

[aem-code-sync]

Hello, I'm the AEM Code Sync Bot and I will run some actions to deploy your branch.
In case there are problems, just click the checkbox below to rerun the respective action.

• Re-sync branch

Commits

• 6363066 ✅ (latest)
• 3b446f5 ✅
• eb0f0e7 ✅

[anfibiacreativa]

we're keying retry-vs-error off a prose prefix from the backend. that breaks silently the day AO changes the wording. is there really no structured signal (a status/error_type field) we can gate on instead? at minimum worth a comment on the AO side that this string is now a contract.

[sharanyavinod]

This is just in the interim until we figure out what exactly is causing the issue. And in the end all this does on the client side is swap out the label to not show ERROR explicitly - all communication stays the same and responses and still intact when expanded
AO does the retry itself, not on demand by the UI - our changes are purely cosmetic

[sharanyavinod]

Maybe we can avoid status in the tool call altogether so we can sidestep this whole concern atleast from client perspective. Coworker does something similar - it has the same issues, just not very visible to the user. For this I would like to discuss with Kilian, so might the current solution make sense in the interim? Wdyt?