test(e2e): implement harness skeleton + synthetic smoke test + pytest markers

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

pyproject.toml

@@ -52,3 +52,8 @@ select = ["E", "F", "I", "W"]
 [tool.pytest.ini_options]
 testpaths = ["tests"]
 asyncio_mode = "auto"
+markers = [
+    "e2e: end-to-end test against a real dataset",
+    "e2e_slow: e2e test that takes > 2 minutes, nightly-only",
+    "needs_dataset: test requires an external dataset to be downloaded",
+]