fix: keep UpdateCheckWorker alive until its thread finishes (#32)

Closing the About dialog mid-check GC'd the dialog and the running
QThread with it -> 'QThread: Destroyed while thread is still running'.
A class-level keepalive set now holds each worker until finished;
stale results to a destroyed receiver are dropped by Qt.

Closes #32

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

bcc.py

@@ -13,6 +13,7 @@ from __future__ import annotations
 import sys
 import time
 from pathlib import Path
+from typing import ClassVar
 
 from PySide6.QtCore import QRect, QSettings, QSize, Qt, QThread, QTimer, QUrl, Signal
 from PySide6.QtGui import (
@@ -172,10 +173,30 @@ class UpdateCheckWorker(QThread):
     running binary. Fails quiet — emits None on any network problem — so
     it's safe to fire unattended from a silent startup check as well as from
     the About dialog's "Check for updates" button.
+
+    Lifetime: the class keeps every instance alive in `_live` until its
+    thread has finished. Without this, a caller whose own reference dies
+    early (the About dialog is a temporary — closing it mid-check used to
+    GC the dialog and the running QThread with it) crashes the process with
+    "QThread: Destroyed while thread is still running". Callers may drop
+    their reference at any time; signal connections to a destroyed receiver
+    are disconnected by Qt, so a late result is simply discarded.
     """
 
     done = Signal(object)  # dict | None
 
+    _live: ClassVar[set[UpdateCheckWorker]] = set()
+
+    def __init__(self):
+        super().__init__()
+        UpdateCheckWorker._live.add(self)
+        self.finished.connect(self._release_keepalive)
+
+    def _release_keepalive(self):
+        # Delivered on the main thread after run() has returned; only now is
+        # it safe for the last reference to drop.
+        UpdateCheckWorker._live.discard(self)
+
     def run(self):
         self.done.emit(core.fetch_latest_release())