fix(call-to-play): compact expired snapshots

Expired call histories were removed on the next store insertion, so an otherwise
idle peer could continue carrying stale payload through handshakes after the
five-minute UI retention ended.

Run the same inactive-call compaction before local and handshake snapshots. This
makes the expiry boundary exact without trimming any event from an active call.

Test Plan:
- `just fmt` -- passed
- `just clippy` -- passed
- `just test` -- passed, 147 peer tests
- `git diff --cached --check` -- passed

crates/lanspread-peer/src/call_to_play.rs

@@ -30,7 +30,12 @@ pub(crate) struct CallToPlayStore {
 }
 
 impl CallToPlayStore {
-    pub(crate) fn snapshot(&self) -> Vec<CallToPlayEvent> {
+    pub(crate) fn snapshot(&mut self) -> Vec<CallToPlayEvent> {
+        self.snapshot_at(now_ms())
+    }
+
+    fn snapshot_at(&mut self, now: i64) -> Vec<CallToPlayEvent> {
+        self.compact_inactive_calls(now);
         self.events.clone()
     }
 
@@ -434,8 +439,6 @@ mod tests {
             .insert(create_event("create"))
             .expect("active call should fit");
 
-        store.compact_inactive_calls(TEST_NOW + 5 * 60_000 + 60_001);
-
-        assert!(store.snapshot().is_empty());
+        assert!(store.snapshot_at(TEST_NOW + 5 * 60_000 + 60_001).is_empty());
     }
 }