perf(daemon): memoize the external debug.log miss

When following the debug.log of a node this session did not launch, every
poll re-read the last 2 MB of the file and re-scanned it for the process
banner. If that node logs to its console instead, the banner is never
there, so the full re-scan repeated on every poll for the life of the
process.

- A miss is now remembered by pid and file size: if nothing has changed
  since the last miss, or the file is growing but the backoff has not
  elapsed, the scan is skipped.
- The retry schedule re-scans quickly during a fresh-node grace window
  (the banner flushes a beat after the pid appears) and then backs off
  from 4 s, doubling to a 60 s cap. The memo is cleared on a successful
  tie and when the followed process goes away.

Tests: the retry delay is fast within the grace window and then steps
4, 8, 16 s up to the 60 s cap.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

src/daemon/daemon_busy_detect.h

@@ -141,6 +141,17 @@ inline bool shouldConfirmStop(bool stopsDaemon, const DaemonBusyTracker& tracker
     return stopsDaemon && (tracker.rebuilding() || rpcBusyWhileAlive);
 }
 
+// Retry schedule after failing to find a pid's banner in debug.log: re-scan fast during the fresh-node grace
+// window, then back off 4 s doubling to a 60 s cap. `backoff` is the caller's running state (0 until grace ends).
+inline double nextExternalLogRetryDelay(double ageSinceMiss, double& backoff,
+                                        double graceWindow = 15.0, double fast = 2.0,
+                                        double base = 4.0, double cap = 60.0)
+{
+    if (ageSinceMiss < graceWindow) return fast;
+    backoff = (backoff < base) ? base : (backoff * 2.0 < cap ? backoff * 2.0 : cap);
+    return backoff;
+}
+
 /**
  * Where to start feeding a DaemonBusyTracker from the last `window` of the debug.log of a node someone else
  * started at `processStartUtc` (UTC seconds). Only lines that process wrote count: a node started with