fix(proto): tolerate undecodable peer datagrams instead of disconnecting

The DTLS recv loops in src/listen.rs and src/connect.rs each read
one full datagram per call. A failed `try_into::<ProtoEvent>()`
means the datagram's leading EventType byte didn't match any
known variant — a misalignment is impossible because DTLS is
message-framed, not stream-framed.

Previously, src/listen.rs would `break` out of the loop on parse
failure (tearing down the connection) and src/connect.rs would
silently swallow the error with no log. Both are wrong as
forward-compat behavior: any future protocol addition (e.g. a
new event variant) would force every existing peer to disconnect
rather than gracefully ignoring the unknown event.

Skip-and-continue on both sides, with a debug-level log so the
behavior is observable. Pre-requisite for any future ProtoEvent
variant to land without forcing a coordinated upgrade across
every peer in a deployment.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/connect.rs

@@ -255,16 +255,23 @@ async fn receive_loop(
 ) {
     let mut buf = [0u8; MAX_EVENT_SIZE];
     while conn.recv(&mut buf).await.is_ok() {
-        if let Ok(event) = buf.try_into() {
-            log::trace!("{addr} <==<==<== {event}");
-            match event {
-                ProtoEvent::Pong(b) => {
-                    client_manager.set_active_addr(handle, Some(addr));
-                    client_manager.set_alive(handle, b);
-                    ping_response.borrow_mut().insert(addr);
+        match buf.try_into() {
+            Ok(event) => {
+                log::trace!("{addr} <==<==<== {event}");
+                match event {
+                    ProtoEvent::Pong(b) => {
+                        client_manager.set_active_addr(handle, Some(addr));
+                        client_manager.set_alive(handle, b);
+                        ping_response.borrow_mut().insert(addr);
+                    }
+                    event => tx.send((handle, event)).expect("channel closed"),
                 }
-                event => tx.send((handle, event)).expect("channel closed"),
             }
+            // Skip undecodable datagrams without dropping the
+            // connection. Each DTLS recv is one framed message, so
+            // skipping is safe and keeps us forward-compatible with
+            // peers that send event types we don't yet know about.
+            Err(e) => log::debug!("ignoring undecodable event from {addr}: {e}"),
         }
     }
     log::warn!("recv error");

src/listen.rs

@@ -259,8 +259,16 @@ async fn read_loop(
                 .send(ListenEvent::Msg { event, addr })
                 .expect("channel closed"),
             Err(e) => {
-                log::warn!("error receiving event: {e}");
-                break;
+                // Skip the malformed/unknown datagram and keep
+                // listening. Each DTLS recv returns one full
+                // datagram, so a parse error here can't desync a
+                // stream; the next call gets a fresh, framed
+                // message. This makes the protocol forward-
+                // compatible: a peer running a newer Lan Mouse
+                // version can introduce additional event types
+                // and old peers will simply ignore them rather
+                // than dropping the connection.
+                log::debug!("ignoring undecodable event from {addr}: {e}");
             }
         }
     }