search: upload the executor command once, not on every poll

ObjectStoreExecutor.submit re-PUT the identical command.json on every
0.2s result poll, so a task running minutes wrote ~1,500 versions of one
engine-queue/search-exec/<id>/command.json. On the versioned memory
bucket that was the dominant growth: 109,715 versions, 10.3 GB (91% of
the bucket). retry_mailbox already retries a lost PUT acknowledgement
internally, so one upload per task is the correct delivery guarantee.

Guard the PUT with an `uploaded` flag: upload once (still retried on
transient failure), then poll only the result. One command.json version
per task instead of ~1,500. All submit semantics preserved.

59 executor tests pass, ruff clean.

Co-Authored-By: Claude Opus 4.8 <[email protected]>

src/openultrasast/search/executor.py

@@ -583,13 +583,16 @@ class ObjectStoreExecutor:
         retry_end = end + RETRY_SECONDS
         last_error = ""
         phase = "result upload"
+        uploaded = False
         while time.monotonic() < end:
             try:
-                # A lost PUT acknowledgement is retried with the exact same sequence.
-                phase = "command upload"
                 before = time.monotonic()
                 try:
-                    retry_mailbox(lambda timeout: self.transport.put(self.command_url, payload, timeout), deadline=retry_end)
+                    if not uploaded:
+                        # Retry lost acknowledgements internally; successful delivery needs no further PUTs.
+                        phase = "command upload"
+                        retry_mailbox(lambda timeout: self.transport.put(self.command_url, payload, timeout), deadline=retry_end)
+                        uploaded = True
                     phase = "result read"
                     raw = retry_mailbox(lambda timeout: self.transport.get(self.result_url, timeout), deadline=retry_end)
                 finally: