Move ChannelMonitor state behind a Mutex

ChainMonitor accesses a set of ChannelMonitors behind a single Mutex.
As a result, update_channel operations cannot be parallelized. It also
requires using a RefCell around a ChannelMonitor when implementing
chain::Listen.

Moving the Mutex into ChannelMonitor avoids these problems and aligns it
better with other interfaces. Note, however, that get_funding_txo and
get_outputs_to_watch now clone the underlying data rather than returning
references.

Author: jkczyz

fuzz/src/chanmon_consistency.rs

@@ -126,7 +126,7 @@ impl chain::Watch<EnforcingSigner> for TestChainMonitor {
 			hash_map::Entry::Occupied(entry) => entry,
 			hash_map::Entry::Vacant(_) => panic!("Didn't have monitor on update call"),
 		};
-		let mut deserialized_monitor = <(BlockHash, channelmonitor::ChannelMonitor<EnforcingSigner>)>::
+		let deserialized_monitor = <(BlockHash, channelmonitor::ChannelMonitor<EnforcingSigner>)>::
 			read(&mut Cursor::new(&map_entry.get().1), &OnlyReadsKeysInterface {}).unwrap().1;
 		deserialized_monitor.update_monitor(&update, &&TestBroadcaster{}, &&FuzzEstimator{}, &self.logger).unwrap();
 		let mut ser = VecWriter(Vec::new());