fix: suppress Pulsar C++ client log noise (#936)

Revert consumer receive timeout from 100ms back to the original
2000ms.  The 100ms change was based on a misunderstanding — receive()
is a blocking call that returns immediately when a message arrives,
so the timeout only affects how quickly a consumer checks the shutdown
flag during idle periods.  100ms generated ~200 WARN lines/sec from
the C++ client with no latency benefit.

Also set the Pulsar C++ client logger to Error level so residual
timeout warnings from the subscriber (250ms) don't produce noise.

Update poll timeout test to match reverted 2000ms value

trustgraph-base/trustgraph/base/consumer.py

@@ -188,7 +188,7 @@ class Consumer:
             try:
                 msg = await loop.run_in_executor(
                     executor,
-                    lambda: consumer.receive(timeout_millis=100),
+                    lambda: consumer.receive(timeout_millis=2000),
                 )
             except Exception as e:
                 # Handle timeout from any backend

trustgraph-base/trustgraph/base/pulsar_backend.py

@@ -139,6 +139,10 @@ class PulsarBackend:
         if api_key:
             client_args['authentication'] = pulsar.AuthenticationToken(api_key)
 
+        client_args['logger'] = pulsar.ConsoleLogger(
+            _pulsar.LoggerLevel.Error
+        )
+
         self.client = pulsar.Client(**client_args)
         logger.info(f"Pulsar client connected to {host}")