Test run errors from xctest runner failures now appear as target-level Errors instead of fake test cases

When xctest itself fails (for example, a target’s .xctest bundle can’t be loaded or the app can’t launch), the processor now classifies the failure as a run or target-level Error instead of a synthetic test case. The run detail page shows these errors in a dedicated Errors card above the renamed Test Failures section. This stops the previous flood of test_case.created webhooks and test-count inflation caused by each process-id producing a distinct fake test case, and gives test runs a clearer status.