fix: add loopback to connector bind addresses for localhost connections

set_bind_addr_for_peer_connector collected all local interface IPs
as bind addresses but omitted 127.0.0.1. When connecting to localhost,
the connector would bind to a non-loopback IP (e.g. 172.17.0.2 in
Docker) and fail with connect timeout because routing from non-loopback
to loopback doesn't work.

Fix: prepend 127.0.0.1:0 to the bind address list. The connector tries
all bind addresses, so loopback will be attempted first and succeed for
localhost connections.

Benchmark results in Docker (4 threads, 1400B, 10s):
  Ring: 508K pps, MpscTunnelSender::send 138ns
  UDP:  440K pps, MpscTunnelSender::send 294ns
  TCP:  440K pps, MpscTunnelSender::send 376ns

All three tunnel types now converge and benefit from noop_waker sync send.

scripts/setup-bench-netns.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+set -e
+
+NS_NAME="et_bench"
+
+echo "=== Creating netns: $NS_NAME ==="
+# Clean up old ns
+sudo ip netns del "$NS_NAME" 2>/dev/null || true
+
+# Create namespace
+sudo ip netns add "$NS_NAME"
+
+# Enable loopback inside namespace
+sudo ip netns exec "$NS_NAME" ip link set lo up
+
+echo "=== netns $NS_NAME ready ==="
+echo "Both instances will share this namespace's loopback."
+echo "TCP/UDP connections to 127.0.0.1 will work inside it."
+echo ""
+echo "Now run the bench with HOTPATH_NETNS=$NS_NAME:"
+echo "  HOTPATH_TUNNEL=tcp HOTPATH_NETNS=$NS_NAME HOTPATH_BENCH_SECS=15 \\"
+echo "    cargo run --profile hotpath --features hotpath --example cpu_hotspot_ring"