fix(frontend): improve error message extraction in agent execution error handler (#11527)

When agent execution fails, the error toast was only showing
`error.message`, which often lacks detail. The API returns more specific
error messages in `error.response.detail.message`, but these weren't
being displayed to users, making debugging harder.

<img width="1018" height="796" alt="Screenshot 2025-12-03 at 2 14 17 PM"
src="https://github.com/user-attachments/assets/6a93aed9-9e18-450a-9995-8760d7d5ca35"
/>

### Changes 🏗️

- Updated error message extraction in `useAgentRunModal` to check
`error.response.detail.message` first, then fall back to
`error.message`, then to the default message
- This ensures users see the most specific error message available from
the API response

### Checklist 📋

#### For code changes:

- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Triggered an agent execution error (e.g., invalid inputs) and
verified the toast shows the detailed error message from
`error.response.detail.message`
- [x] Verified fallback to `error.message` when
`error.response.detail.message` is not available
  - [x] Verified fallback to default message when neither is available

autogpt_platform/frontend/src/app/(platform)/library/agents/[id]/components/NewAgentLibraryView/components/modals/RunAgentModal/useAgentRunModal.ts

@@ -87,9 +87,13 @@ export function useAgentRunModal(
         }
       },
       onError: (error: any) => {
+        const errorMessage = error.isGraphValidationError()
+          ? error.response.detail.message
+          : error.message;
+
         toast({
           title: "❌ Failed to execute agent",
-          description: error.message || "An unexpected error occurred.",
+          description: errorMessage || "An unexpected error occurred.",
           variant: "destructive",
         });
       },