perf(mge/module): optimize conv_bn qat module to improve performance

GitOrigin-RevId: 9415b83d9c

imperative/python/megengine/module/qat/conv_bn.py

@@ -132,13 +132,7 @@ class _ConvBnActivation2d(Float._ConvBnActivation2d, QATModule):
         if self.conv.bias is not None:
             orig_conv = orig_conv + self.conv.bias
         # calculate batch norm
-        bn_mean, bn_var = self.get_batch_mean_var(orig_conv)
-        bn_istd = 1.0 / sqrt(bn_var + self.bn.eps)
-        conv = gamma * bn_istd * (orig_conv - bn_mean) + beta
-        num_elements_per_channel = conv.size / conv.shape[1]
-        self.update_running_mean_and_running_var(
-            bn_mean, bn_var, num_elements_per_channel
-        )
+        conv = self.bn(orig_conv)
         return conv
 
     @classmethod