Turn Nagle's algorithm off on the web server's connections (#170)

axum 0.7 set TCP_NODELAY on every accepted connection; 0.8 stopped, and serve
never set it, so each response sent in parts held its last small segment
until the browser's delayed ACK came back. On production's light traffic
nothing showed: small responses go out in one write. Under load it did.
tests/load, Nagle on then off: media range requests median 41.8 ms to
10.3 ms, p95 44.4 to 13.2, throughput 39 to 149 MB/s. Browse and sign-in
unchanged within noise; flag saves vary 80-195 ms either way, behind the
scan's writes, not the network.

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

src/web.rs

@@ -91,9 +91,17 @@ pub fn router(state: WebState) -> Router {
 }
 
 pub async fn serve(state: WebState, bind: &str) -> Result<()> {
+    use axum::serve::ListenerExt;
+    // Nagle off, as axum 0.7 had it: under load a response goes out in parts, and with Nagle
+    // on its last small part waits for the browser's delayed ACK, up to 40 ms (#170).
     let listener = tokio::net::TcpListener::bind(bind)
         .await
-        .with_context(|| format!("binding {bind}"))?;
+        .with_context(|| format!("binding {bind}"))?
+        .tap_io(|tcp| {
+            if let Err(e) = tcp.set_nodelay(true) {
+                tracing::debug!(error = %e, "could not set TCP_NODELAY");
+            }
+        });
     tracing::info!(bind, "web ui listening");
     axum::serve(
         listener,