polygon/sync: retry ufc on busy response

[taratorio]

a user on discord flagged the following issue (discord link):

[INFO] [01-31|10:44:40.015] [sync] update fork choice done           in=807.13624ms
[INFO] [01-31|10:44:40.022] [sync] update fork choice                block=67365147 hash=0xb697ef9de67d61f300fda27367617ec270d9bcd15e9a7c73facdc253b6fc0273 age=3s
[EROR] [01-31|10:44:40.022] failed to update fork choice             latestValidHash=0x0000000000000000000000000000000000000000000000000000000000000000 err="fork choice update failure: status=5, validationErr=''"
[EROR] [01-31|10:44:40.038] runPostForkchoiceInBackground            error="[1/6 OtterSync] db closed"
[EROR] [01-31|10:44:43.029] polygon sync crashed - stopping node     err="pos sync failed: fork choice update failure: status=5, validationErr=''"

Note: status=5 means ExecutionStatus_Busy

This can happen because there is a possibility that runPostForkchoiceInBackground acquires the execution engine semaphore after a successful UFC (line which says update fork choice done in=807.13624ms) to do pruning. But very shortly after that there is another call to UFC (in this case 7ms after - line which says update fork choice block=67365147) which then fails with a Busy response since it can't acquire the semaphore (take by runPostForkchoiceInBackground for pruning).

This PR fixes this situation by adding retries on Busy response when calling UFC. Such a retry already existed when calling InsertBlocks so this is something that just got missed.