Re-check that the connection is still stale before we reap it

A concurrent thread may have also detected it to be stale, and already
released (or even reassigned) it by now.

Fixes #25585

activerecord/CHANGELOG.md

+*   Ensure concurrent invocations of the connection reaper cannot allocate the
+    same connection to two threads.
+
+    Fixes #25585.
+
+    *Matthew Draper*
+
 *   Inspecting an object with an associated array of over 10 elements no longer
     truncates the array, preventing `inspect` from looping infinitely in some
     cases.

activerecord/lib/active_record/connection_adapters/abstract/connection_pool.rb

@@ -561,6 +561,8 @@ def reap
 
         stale_connections.each do |conn|
           synchronize do
+            next unless conn.in_use? && !conn.owner.alive?
+
             if conn.active?
               conn.reset!
               checkin conn