Count traffic per client address and report the busiest ones with the port totals

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

README.md

@@ -33,6 +33,7 @@ Settings and secrets can live in a `.env` file instead of the command line; see
 - **Ports:** TCP, UDP, HTTP, HTTPS and TLS passthrough, one listener per IP and port, opened and closed live as you change them in the dashboard. Port ranges for TCP and UDP. Round-robin load balancing and failover across several targets, with health checks.
 - **PROXY protocol (v1/v2)** on TCP and TLS ports, so the target sees the real client address.
 - **Web:** automatic Let's Encrypt certificates, HTTP/2, websockets, and logins (account, password, PIN, emailed code, share links, path and IP rules) enforced in front of any site. Use the built-in proxy, or `-traefik` to let Traefik serve http/https (raw TCP, UDP and TLS-passthrough ports always stay with Blacktip, since Traefik can't open ports without a restart).
+- **Traffic analytics:** counts connections and bytes in and out for every port, and the busiest client addresses on each (the ten biggest per report, at most a few thousand tracked between reports), and sends them to Hammerhead with its regular report. Hammerhead keeps the history, draws the charts and raises traffic alerts. Nothing is stored on this machine, and a report that fails to arrive is added to the next one.
 - **Mail egress:** delivers your mail server's outgoing mail straight to the recipient's server from a public IP you choose, with the HELO name you set, opportunistic STARTTLS, optional DKIM signing, MTA-STS enforcement, a per-minute send limit and a deliverability checklist.
 - **Tunnel:** WireGuard through a userspace network stack, so no root, no kernel interface and no driver. It works on Windows, macOS and Linux, and sites behind NAT connect out to it.