SplitMultipleTapes test segfaults non-deterministically

[erick-xanadu]

https://github.com/PennyLaneAI/catalyst/actions/runs/12605094878 (select Attempt #1)

[paul0403]

It's not a segfault, but a

Assertion failed: (inserted.second && "expected region to contain uniquely named symbol operations")

Maybe something wrong during renameToUnique (or whatever it was called)? I can take a look when I come back...

Edit: I will add though, Romain and I tried this test locally a long time ago and didn't see any problems in like 2 million runs, so very likely this is a race condition that only occurs with git runners?