fix(logging): resolve logging_compat keyword arg conflicts

- Fix logging_compat._log() to handle extra keyword arguments properly
- Update config.py, main.py, and test_bans.py for compatibility
- Update Tasks.md and runner.csx

Docs/Tasks.md

@@ -1,13 +1,3 @@
-# Failed Tests
-
-Total unique failed/errored tests: 406
-
-## 1. TestGetActiveBans.test_401_when_unauthenticated
-
-**Exception:** pydantic_core._pydantic_core.ValidationError: 2 validation errors for Settings
-
----
-
 ## 2. TestGetActiveBans.test_empty_when_no_bans
 
 **Exception:** pydantic_core._pydantic_core.ValidationError: 2 validation errors for Settings
@@ -2434,7 +2424,4 @@ Total unique failed/errored tests: 406
 
 ## 406. TestConfigModuleIntegration.test_filter_config_service_list_filters_uses_imports
 
-**Exception:** AttributeError: module 'app.models.config' has no attribute 'get_settings'
-
----
-
+**Exception:** AttributeError: module 'app.models.config' has no attribute 'get_settings'

Docs/runner.csx

@@ -102,7 +102,7 @@ for (int i = 0; i < items.Count; i++)
 
     // Step 1 — run the task prompt
     await RunCopilot(Enumerable.Empty<string>(), $"/caveman full");
-    await RunCopilot(new[] { "--continue" }, $"read ./Docs/Instructions.md. {item}");
+    await RunCopilot(new[] { "--continue" }, $"read ./Docs/Instructions.md. fix the following test and only that one {item}");
     if (cts.IsCancellationRequested) break;
 
     // Step 2 — confirm completion in the same chat session