test(sync): wait for SIGINT listener registration in listener-baseline test

Same flake class as the SIGHUP fix — "returns SIGINT/SIGTERM listener
counts to their baseline after shutdown" (line 943) surfaced on Node 22
CI once the earlier SIGHUP flake was fixed. Under worker contention the
fixed-50 flushWatchSetup() rounds occasionally return before
watchAndSync reaches its process.on('SIGINT', ...) call, producing
"Expected: 1, Received: 0" at the listenerCount assertion.

Apply the same bounded wait-until-registered loop in both halves of the
test (firstRun and secondRun) so failures remain loud but timing-robust.

Author: sjsyrek

tests/unit/sync/sync-command.test.ts

@@ -953,6 +953,13 @@ describe('SyncCommand', () => {
       // signal to the real process emitter (listeners only react synchronously).
       const firstRun = command.run({ watch: true, debounce: 50 } as CliSyncOptions);
       await flushWatchSetup();
+      // Under heavy CI worker contention the fixed-round flush occasionally
+      // returns before watchAndSync has reached its process.on('SIGINT', ...)
+      // call. Loop with a hard ceiling so we fail loudly rather than with a
+      // misleading "Expected: 1, Received: 0".
+      for (let i = 0; i < 10 && process.listenerCount('SIGINT') === sigintBaseline; i++) {
+        await flushWatchSetup();
+      }
       expect(process.listenerCount('SIGINT')).toBe(sigintBaseline + 1);
       expect(process.listenerCount('SIGTERM')).toBe(sigtermBaseline + 1);
 
@@ -965,6 +972,9 @@ describe('SyncCommand', () => {
       // Second invocation should also add exactly one listener, then remove it.
       const secondRun = command.run({ watch: true, debounce: 50 } as CliSyncOptions);
       await flushWatchSetup();
+      for (let i = 0; i < 10 && process.listenerCount('SIGINT') === sigintBaseline; i++) {
+        await flushWatchSetup();
+      }
       expect(process.listenerCount('SIGINT')).toBe(sigintBaseline + 1);
       expect(process.listenerCount('SIGTERM')).toBe(sigtermBaseline + 1);