Cache.get(...) using Flag.FORCE_WRITE_LOCK should retry on OutdatedTopologyException

From IRC:
pferraro: Following a shutdown of a node, we're seeing OutdatedTopologyExceptions during a Cache.get(...) using Flag.FORCE_WRITE_LOCK when the requested key is not owned by the requesting node
pferraro: is there a reason why Infinispan doesn't automatically retry here?
dberindei: I think we just overlooked it
dberindei: we are retrying a LockControlCommand when you call lock() explicitly
dberindei: but maybe not when it's invoked for a get()
pferraro_: is that something that can be fixed easily?
dberindei: yes, I think so