search-with-proof: diagnosable failures, pip install race, no duplicate tasks

Reasons carry type plus a sanitised message tail; command failures carry phase,
exit code and bounded stderr. Scratch walk tolerates directories pip removes
mid-walk; writable /workspace/home and runtime paths. Capacity retries delete a
possibly created task before resubmitting. Docker smoke checks the image first.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

tests/test_search_worker.py

@@ -113,7 +113,7 @@ def test_reservation_precedes_each_call_and_refuses_over_budget(tmp_path):
     job = task()
     job.limits["max_call_usd"] = 0.000001
     result = worker(job)
-    assert result["failure"] == "BudgetExhausted"
+    assert result["failure"].startswith("BudgetExhausted: ")
     assert len(client.calls) == 1