fix: drop pam_acct_mgmt from password and FP paths (v0.6.13)

The PAM stack only ever had `auth include login` — no account module.
auth.rs nevertheless called pam_acct_mgmt after pam_authenticate, which
fell back to /etc/pam.d/other (pam_deny) and rejected every password.
On the FP side, the same call was wrapped in a spawn_blocking + 2s
resume_async retry path that triggered a use-after-free in
gtk_window_destroy (20+ SIGSEGVs in 6 days).

- auth.rs: remove pam_acct_mgmt extern + call; return pam_authenticate
  result directly. Lockout still works via pam_faillock in the auth stack.
- auth.rs: drop check_account() and its tests (FP path no longer needs it).
- lockscreen.rs::start_fingerprint: on success go straight to
  label.set_text + fp.stop() + cb(); no PAM acct check, no resume retry.
- fingerprint.rs: remove resume_async() — no caller left.
- config/moonlock-pam: keep single `auth include login` line, matching
  swaylock/gtklock pattern.
- CLAUDE.md, DECISIONS.md updated.

src/fingerprint.rs

@@ -175,29 +175,6 @@ impl FingerprintListener {
         Self::begin_verification(listener, username).await;
     }
 
-    /// Resume fingerprint verification after a transient interruption (e.g. failed
-    /// PAM account check). Reuses previously stored callbacks. Re-claims the device
-    /// and restarts verification from scratch. Awaits any in-flight VerifyStop +
-    /// Release before re-claiming the device so fprintd does not reject the Claim
-    /// while the previous session is still being torn down.
-    pub async fn resume_async(
-        listener: &Rc<RefCell<FingerprintListener>>,
-        username: &str,
-    ) {
-        // Drain in-flight cleanup so the device is actually released before Claim.
-        // Without this, a fast resume after on_verify_status's fire-and-forget
-        // cleanup races the Release call and fprintd returns "already claimed".
-        let proxy = listener.borrow_mut().take_cleanup_proxy();
-        if let Some(proxy) = proxy {
-            Self::perform_dbus_cleanup(proxy).await;
-        }
-        // Deliberately do NOT reset failed_attempts here. An attacker with sensor
-        // control could otherwise cycle verify-match → check_account fail → resume,
-        // and the 10-attempt cap would never trigger. The counter decays only via
-        // a fresh lock session (listener construction).
-        Self::begin_verification(listener, username).await;
-    }
-
     /// Claim device, start verification, and connect D-Bus signal handler.
     /// Assumes device_proxy is set and callbacks are already stored.
     async fn begin_verification(
@@ -366,11 +343,6 @@ impl FingerprintListener {
 
     /// Disconnect the signal handler and clear running flags. Returns the proxy
     /// the caller should use for the async D-Bus cleanup (VerifyStop + Release).
-    ///
-    /// Split into a sync part (signal disconnect, flags) and an async part
-    /// (`perform_dbus_cleanup`) so callers can either spawn the async work
-    /// fire-and-forget (via `cleanup_dbus`) or await it to serialize with a
-    /// subsequent Claim (via `resume_async`).
     fn take_cleanup_proxy(&mut self) -> Option<gio::DBusProxy> {
         self.running = false;
         self.running_flag.set(false);