iommu/arm-smmu: use a threaded handler for global faults

We need to do some sleeping operations (like enabling clocks) in the
global fault handler.  Move to a threaded handler to avoid BUGs due to
sleeping in an atomic context.

Change-Id: I88eed743bfad4cc33daf3b9f81255587c72b0167
Signed-off-by: Mitchel Humpherys <mitchelh@codeaurora.org>