service: storage_proxy: make hint write handlers cancellable

Whether a write handler should be cancellable is now controlled by a
parameter passed to `create_write_response_handler`. We plumb it down
from `send_to_endpoint` which is called by hints manager.

This will cause hint write handlers to immediately timeout when we
shutdown or when a destination node is marked as dead.

Fixes #8079

db/view/view.cc

@@ -1605,7 +1605,8 @@ static future<> apply_to_remote_endpoints(service::storage_proxy& proxy, gms::in
             std::move(pending_endpoints),
             db::write_type::VIEW,
             std::move(tr_state),
-            allow_hints);
+            allow_hints,
+            service::is_cancellable::yes);
 }
 
 static bool should_update_synchronously(const schema& s) {