Add MSE missing operators

Context

In order to use MSE with backend cuda, the following operators need to be supported:

• Mul
• ReduceMean
• Pow

Solves #20 (closed)

Modified files

• MulImpl.hpp, MulImpl.cpp and Test_MulImpl.cpp, add Mul forward() and backward();
• ReduceMeanImpl.hpp, ReduceMeanImpl.cpp and Test_ReduceMeanImpl.cpp, add ReduceMean forward() and backward();
• PowImpl.hpp, PowImpl.cpp and Test_PowImpl.cpp, add Pow forward() and backward();

TODO

• Mul forwad
• Mul backward
• ReduceMean forward
• ReduceMean backward
• Pow forward

include/aidge/backend/cuda/operator/MulImpl.hpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#ifndef AIDGE_BACKEND_CUDA_OPERATOR_MULIMPL_H_
+#define AIDGE_BACKEND_CUDA_OPERATOR_MULIMPL_H_
+
+#include <array>
+#include <memory>
+#include <tuple>
+#include <vector>
+
+#include <cudnn.h>
+
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/operator/Mul.hpp"
+#include "aidge/utils/Registrar.hpp"
+#include "aidge/utils/Types.h"
+
+#include "aidge/backend/cuda/utils/CudaUtils.hpp"
+
+namespace Aidge {
+class MulImpl_cuda : public OperatorImpl {
+private:
+
+public:
+    MulImpl_cuda(const Mul_Op &op) : OperatorImpl(op, "cuda") {}
+
+    static std::unique_ptr<MulImpl_cuda> create(const Mul_Op &op) {
+        return std::make_unique<MulImpl_cuda>(op);
+    }
+
+public:
+    void forward();
+    void backward();
+
+private:
+    template <class T> void forward_(const std::vector<Tensor>& inputs, const std::vector<std::vector<int>>& inputsDims, const std::vector<std::vector<int>>& inputsStrides);
+    template <class T> void backward_(const Tensor& outputGrad, const std::vector<std::vector<int>>& inputsDims, const std::vector<std::vector<int>>& inputsStrides);
+};
+
+namespace {
+// add cuda backend to Mul_Op implementation registry
+static Registrar<Mul_Op> registrarMulImpl_cuda("cuda", Aidge::MulImpl_cuda::create);
+}  // namespace
+}  // namespace Aidge
+
+#endif /* AIDGE_BACKEND_CUDA_OPERATOR_MULIMPL_H_ */