fix index_put bug when index is multi-dim bool tensor (#55191)

* fix index_put bug when index is multi-dim bool tensor

* fix name error

paddle/phi/kernels/cpu/index_put_grad_kernel.cc

@@ -195,7 +195,7 @@ void IndexPutGradKernel(const Context& dev_ctx,
   std::vector<DenseTensor> tmp_res_indices_v;
   std::vector<DenseTensor> range_tensor_v;
 
-  for (int i = indices.size(); i < x.dims().size(); ++i) {
+  for (int i = int_indices_v.size(); i < x.dims().size(); ++i) {
     range_tensor_v.emplace_back(funcs::GetRangeTensor<int64_t, Context>(
         dev_ctx, x.dims()[i], phi::DataType::INT64));
   }

paddle/phi/kernels/cpu/index_put_kernel.cc

@@ -127,7 +127,7 @@ void IndexPutKernel(const Context& dev_ctx,
   std::vector<DenseTensor> range_tensor_v;
   const DenseTensor* ptr_value = nullptr;
 
-  for (int i = indices.size(); i < x.dims().size(); ++i) {
+  for (int i = int_indices_v.size(); i < x.dims().size(); ++i) {
     range_tensor_v.emplace_back(funcs::GetRangeTensor<int64_t, Context>(
         dev_ctx, x.dims()[i], phi::DataType::INT64));
   }

paddle/phi/kernels/gpu/index_put_grad_kernel.cu

@@ -227,7 +227,7 @@ void IndexPutGradKernel(const Context& dev_ctx,
   std::vector<DenseTensor> tmp_res_indices_v;
   std::vector<DenseTensor> range_tensor_v;
 
-  for (int i = indices.size(); i < x.dims().size(); ++i) {
+  for (int i = int_indices_v.size(); i < x.dims().size(); ++i) {
     range_tensor_v.emplace_back(funcs::GetRangeCudaTensor<int64_t, Context>(
         dev_ctx, x.dims()[i], phi::DataType::INT64));
   }

paddle/phi/kernels/gpu/index_put_kernel.cu

@@ -128,7 +128,7 @@ void IndexPutKernel(const Context& dev_ctx,
   std::vector<DenseTensor> range_tensor_v;
   const DenseTensor* ptr_value = nullptr;
 
-  for (int i = indices.size(); i < x.dims().size(); ++i) {
+  for (int i = int_indices_v.size(); i < x.dims().size(); ++i) {
     range_tensor_v.emplace_back(funcs::GetRangeCudaTensor<int64_t, Context>(
         dev_ctx, x.dims()[i], phi::DataType::INT64));
   }