Speed up flake8-async by telling libcst to skip deepcopy. (#438)

* Speed up flake8-async by telling libcst to skip deepcopy.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: yilei

flake8_async/runner.py

@@ -138,7 +138,14 @@ def __init__(self, options: Options, module: Module):
 
     def run(self) -> Iterable[Error]:
         for v in (*self.utility_visitors, *self.visitors):
-            self.module = cst.MetadataWrapper(self.module).visit(v)
+            # The default deepcopy guards against the same CST node object
+            # appearing at two positions in the tree (metadata is keyed by node
+            # identity). Parser output and the result of a prior .visit() never
+            # share nodes, so the copy is wasted work. This stays safe as long
+            # as no visitor returns a cached CST node from multiple leave_* calls.
+            self.module = cst.MetadataWrapper(self.module, unsafe_skip_copy=True).visit(
+                v
+            )
 
         yield from self.state.problems