This patch doesn't actually fix the bug I was hoping to fix, which is
that `update_diagnostics()` sometimes gets called even when there are
more updates that should be processed first. I did eventually figure out
that this issue is caused by Tokio's cooperative yielding, which
currently can't be disabled.
However overall it makes the debounce code somewhat more readable IMO,
which is why I'm suggesting to land it anyway.
The LSP code had numerous places where competing threads could take out
out locks in different orders, making it very prone to deadlocks.
This commit sidesteps the entire issue by switching to a single lock.
The above is a little white lie: the Sources struct still uses a mutex
internally to avoid having to boil the ocean (because being honest about
what it does involves changing all its methods to `&mut self` but that
ripples out extensively...) I'll save that one for another day.
