Show xctest runner errors as a separate Errors card on test runs, instead of polluting test cases

Test run pages no longer treat xctest runner errors (for example a target whose .xctest bundle can’t be loaded, or an app that can’t launch) as brand-new test cases. These errors are lifted out and surfaced as a dedicated “Errors” card on the run detail, grouped per target with destructive alert styling and a monospace message block, sitting directly above the test failures section. As a side effect, test counts stay accurate and the flood of per-process test_case.created webhooks and unbounded per-pid test_case rows is stopped at the source. The run page’s failures label is renamed to “Test Failures” to make the distinction from run/target-level Errors obvious. The single test-case-run page keeps the old “Failures” label.