test(ui): poll for the debounced skill auto-save instead of a fixed sleep

Task_toggling_a_skill_auto_saves_selection waited a hard 500 ms for a
debounced save. It passed in isolation and in either half of the suite,
but failed in a full run: this batch added three DB-backed UI test
classes whose real SQLite contexts load the thread pool enough that the
timer callback misses the window. Product code is unchanged.

Same assertion, polled with a 5 s deadline - the pattern the newly added
DetailsIsland/TasksIsland tests already use.

tests/ClaudeDo.Ui.Tests/ViewModels/DetailsIslandTaskUpdatedTests.cs

@@ -79,7 +79,7 @@ public class DetailsIslandTaskUpdatedTests : IDisposable
             new ClaudeDo.Ui.Services.MergeCoordinator());
 
     [Fact]
-    public async Task TaskUpdated_ForBoundTask_RefreshesStatusAndPickUpInTerminal()
+    public async Task TaskUpdated_ForBoundTask_RefreshesStatus()
     {
         await SeedTaskAsync("t-bound-1", TaskStatus.Running);
 
@@ -88,7 +88,7 @@ public class DetailsIslandTaskUpdatedTests : IDisposable
         vm.Bind(new TaskRowViewModel { Id = "t-bound-1", Status = TaskStatus.Running });
         await Task.Delay(50);
 
-        Assert.False(vm.CanPickUpInTerminal);
+        Assert.Equal(TaskStatus.Running, vm.Task?.Status);
 
         await using (var db = NewContext())
         {
@@ -102,7 +102,6 @@ public class DetailsIslandTaskUpdatedTests : IDisposable
             await Task.Delay(25);
 
         Assert.Equal(TaskStatus.WaitingForReview, vm.Task?.Status);
-        Assert.True(vm.CanPickUpInTerminal);
     }
 
     [Fact]