feat: per-listener latency tracking

Each listener now tracks chain setup latency independently via a
dict[str, LatencyTracker] on Metrics. The global aggregate stays for
summary output. /status embeds per-listener latency on each listener
entry; /metrics includes a listener_latency map keyed by host:port.

tests/test_api.py

@@ -135,11 +135,18 @@ class TestHandleStatus:
             ],
         )
         ctx = _make_ctx(config=config)
+        # record some latency for the first listener
+        ctx["metrics"].get_listener_latency("0.0.0.0:1080").record(0.2)
         _, body = _handle_status(ctx)
         assert len(body["listeners"]) == 2
         assert body["listeners"][0]["chain"] == ["socks5://127.0.0.1:9050"]
         assert body["listeners"][0]["pool_hops"] == 0
         assert body["listeners"][1]["pool_hops"] == 1
+        # per-listener latency present on each entry
+        assert "latency" in body["listeners"][0]
+        assert body["listeners"][0]["latency"]["count"] == 1
+        assert "latency" in body["listeners"][1]
+        assert body["listeners"][1]["latency"] is None
 
 
 class TestHandleMetrics:
@@ -156,6 +163,7 @@ class TestHandleMetrics:
         assert "uptime" in body
         assert "rate" in body
         assert "latency" in body
+        assert "listener_latency" in body
 
 
 class TestHandlePool: