Record rate-limit waits in benchmark runs, and leave them out of time

The client waits RATE_LIMIT_RETRY_S and retries once on a 429, silently.
A benchmark run that waited looked slower than the model is: on
2026-09-27, fast models under the unit-test key's 200k tokens/minute
limit had 20-45 s runs that were probably waits.

Both clients now report last_rate_limit_wait_s after every call, even
one that then fails. The benchmark sums it per run into
rate_limit_wait_s, the report's time column leaves it out, and a new
429 column counts the runs that waited. Results stored before this
load unchanged.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

nisse-dev/tests/test_llm.py

@@ -142,6 +142,9 @@ def test_rate_limit_is_retried_once_then_surfaced() -> None:
         client.complete(EVENTS, tools=None, model="local/qwen")
     assert len(attempts) == 2
     assert waited == [RATE_LIMIT_RETRY_S]
+    # Reported even though the call failed, so a caller can leave the wait
+    # out of how long the model took.
+    assert client.last_rate_limit_wait_s == RATE_LIMIT_RETRY_S
 
 
 def test_rate_limit_that_clears_succeeds() -> None:
@@ -149,6 +152,19 @@ def test_rate_limit_that_clears_succeeds() -> None:
     client = make_client(lambda request: responses.pop(0), sleep=lambda _s: None)
     [event] = client.complete(EVENTS, tools=None, model="local/qwen")
     assert isinstance(event, AssistantText)
+    assert client.last_rate_limit_wait_s == RATE_LIMIT_RETRY_S
+
+
+def test_the_rate_limit_wait_is_per_call() -> None:
+    responses = [
+        httpx.Response(429, text="slow down"),
+        completion({"content": "one"}),
+        completion({"content": "two"}),
+    ]
+    client = make_client(lambda request: responses.pop(0), sleep=lambda _s: None)
+    client.complete(EVENTS, tools=None, model="local/qwen")
+    client.complete(EVENTS, tools=None, model="local/qwen")
+    assert client.last_rate_limit_wait_s == 0.0
 
 
 def test_connection_failure_raises_llm_error() -> None: