fix(lite): flush the wallet save on quit before _Exit

On quit the lite controller's destructor never runs, because main()
ends the process with _Exit(). Any state the background refresh had
scanned since the last save was lost: the destructor is where that
final save would normally happen, and it was being skipped on the
one path that always runs.

- shutdown() now calls a new flushOnExit() that persists once on the
  way out, bounded to 4 s so the existing 15 s shutdown watchdog
  budget still holds. On timeout it abandons the save thread and lets
  _Exit follow.
- It is a plain save through the same backend path, which writes a
  temp file and renames it into place. That cannot corrupt the wallet
  if it is cut off or abandoned, and it does not change the wallet
  file format - it only persists sooner on the quit path.
- It reuses the destructor's guards and also skips when a wallet op
  is already in progress: shutdown() has already waited for ops to go
  idle, and a still-pending op holds the sync lock and issues its own
  save, so flushOnExit would only block the budget on a save that
  cannot take the lock.

Tests cover a save on the open, synced, idle path (exactly once), a
blocked save returning within the budget, and the retired and
never-opened guards skipping the save.

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

src/wallet/lite_wallet_controller.cpp

@@ -11,7 +11,11 @@
 
 #include <algorithm>
 #include <chrono>
+#include <condition_variable>
 #include <cstdint>
+#include <memory>
+#include <mutex>
+#include <thread>
 #include <unordered_map>
 #include <utility>
 
@@ -499,6 +503,27 @@ LiteWalletController::LiteWalletController(WalletCapabilities capabilities,
     status_ = lifecycle_.status();
 }
 
+void LiteWalletController::flushOnExit(std::chrono::milliseconds budget)
+{
+    // Same guards as the destructor, plus walletOpInProgress(): App::shutdown() already ran
+    // waitWalletOpsIdle(); if an op is still pending it holds sync_lock and issues its own save,
+    // so skip rather than block the budget on a save that cannot acquire the lock.
+    if (retired_.load() || !walletOpen_.load() || !syncComplete() ||
+        broadcastInProgress() || walletOpInProgress() || !bridge_) return;
+    struct Done { std::mutex m; std::condition_variable cv; bool done = false; };
+    auto st = std::make_shared<Done>();
+    auto bridge = bridge_;                               // shared_ptr copy; never captures `this`
+    try {
+        std::thread([bridge, st]() {
+            bridge->execute("save", "");                 // may block on sync_lock; abandoned on timeout
+            { std::lock_guard<std::mutex> l(st->m); st->done = true; }
+            st->cv.notify_all();
+        }).detach();
+    } catch (...) { return; }                            // no thread: skip (pre-fix behaviour, no crash)
+    std::unique_lock<std::mutex> l(st->m);
+    st->cv.wait_for(l, budget, [&] { return st->done; });  // timeout: abandon the thread; _Exit follows
+}
+
 LiteWalletController::~LiteWalletController()
 {
     scan_->cancel.store(true);   // the detached sync thread stops retrying (it may be in a back-off sleep)
@@ -515,7 +540,8 @@ LiteWalletController::~LiteWalletController()
     }
     // The sync thread may be blocked in an uninterruptible full scan; detach it. It holds
     // shared refs (bridge_ + syncDone_), so it stays safe and the bridge survives until it
-    // finishes — the process is exiting, so a late litelib_shutdown is harmless.
+    // finishes — on the quit path the process is exiting, so a late litelib_shutdown is harmless
+    // (a server switch already shut the bridge down in retire()).
     if (syncThread_.joinable()) syncThread_.detach();
     // Likewise the broadcast thread (send/shield proving): it captures shared refs (bridge +
     // running flag + result slot), never `this`, so detaching is safe.