Update flatten_op_npu.cc

Author: Liu-xiandong

paddle/fluid/operators/flatten_op_npu.cc

@@ -57,7 +57,7 @@ class Flatten2GradNPUKernel : public framework::OpKernel<T> {
 using Tensor = framework::Tensor;
 
 template <typename DeviceContext, typename T>
-class FlattenConntiguousRangeNPUKernel : public framework::OpKernel<T> {
+class FlattenContiguousRangeNPUKernel : public framework::OpKernel<T> {
  public:
   void Compute(const framework::ExecutionContext &ctx) const override {
     auto *X = ctx.Input<Tensor>("X");