[Add] Clip Operator

-Added tensors of empty shape to represent min and max attributes.

-ForwardDims can now copy min and max tensors into the attributes. In case of conflicts, the data inside the tensors takes precedence over the class attributes.

-Created unit test cases to validate this behavior.

-Implemented Python bindings for the class.

See Issue:aidge_quantization#38 (closed)

Main difference with !190 (closed): Changing Min and Max Attributes into OptionalData inputs to make the Operator compatible with ONNX

Add Clip Operator

include/aidge/operator/Clip.hpp

+/********************************************************************************
+ * Copyright (c) 2023 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_CORE_OPERATOR_CLIP_H_
+#define AIDGE_CORE_OPERATOR_CLIP_H_
+
+#include <memory>
+#include <vector>
+#include <limits>
+
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/graph/Node.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+#include "aidge/utils/Registrar.hpp"
+#include "aidge/utils/StaticAttributes.hpp"
+#include "aidge/utils/Types.h"
+
+namespace Aidge {
+
+enum class ClipAttr { Min, Max };
+
+
+class Clip_Op : public OperatorTensor,
+    public Registrable<Clip_Op, std::string, std::function<std::shared_ptr<OperatorImpl>(const Clip_Op&)>> {
+
+public:
+    static const std::string Type;
+private:
+    using Attributes_ = StaticAttributes<ClipAttr, float, float>;
+    template <ClipAttr e> using attr = typename Attributes_::template attr<e>;
+    const std::shared_ptr<Attributes_> mAttributes;
+
+public:
+    Clip_Op() = delete;
+    Clip_Op(float min,float max) : 
+    OperatorTensor(Type, {InputCategory::Data,InputCategory::OptionalData,InputCategory::OptionalData}, 1),
+    mAttributes(std::make_shared<Attributes_>(attr<ClipAttr::Min>(min), attr<ClipAttr::Max>(max)))
+    {}
+    /**
+     * @brief Copy-constructor. Copy the operator attributes and its output tensor(s), but not its input tensors (the new operator has no input associated).
+     * @param op Operator to copy.
+     */
+    Clip_Op(const Clip_Op& op)
+        : OperatorTensor(op),
+          mAttributes(op.mAttributes)
+
+    {
+        if (op.mImpl){
+            SET_IMPL_MACRO(Clip_Op, *this, op.backend());
+        }else{
+            mImpl = nullptr;
+        }
+    }
+
+    /**
+     * @brief Clone the operator using its copy-constructor.
+     * @see Operator::Clip_Op
+     */
+    std::shared_ptr<Operator> clone() const override 
+    {
+        return std::make_shared<Clip_Op>(*this);
+    }
+    bool dimsForwarded() const override final;
+    bool forwardDims(bool allowDataDependency = false) override final;
+    
+    /**
+     * @brief Setter to specify the backend to use
+     */
+    void setBackend(const std::string& name, DeviceIdx_t device = 0) override final;
+    inline std::shared_ptr<Attributes> attributes() const override { return mAttributes; }
+    /**
+    * @brief Getter and Setter for min attribute value
+    * @return float& 
+    */
+    inline float& min() const noexcept { return mAttributes->getAttr<ClipAttr::Min>(); }
+    /**
+    * @brief Getter and Setter for max attribute value
+    * @return float& 
+    */
+    inline float& max() const noexcept { return mAttributes->getAttr<ClipAttr::Max>(); }
+
+    std::set<std::string> getAvailableBackends() const override;
+
+    static const std::vector<std::string> getInputsName(){
+        return {"data_input","min_empty_tensor","max_empty_tensor"};
+    }
+    static const std::vector<std::string> getOutputsName(){
+        return {"data_output"};
+    }
+};
+    /**
+     * @brief The Clip operator is a tensor operator that performs a clipping operation on tensor elements.
+            This class allows limiting tensor values to a specified range, defined by the min 
+            and max parameters (Tensors of empty shapes). Values outside this range are replaced by the corresponding limit values
+            When ‘min’ is greater than ‘max’, the clip operator sets all the ‘input’ values to the value of ‘max’
+    * @param[in] name Name of the node
+    * @param[in] min Min clip value as attribute
+     * @param[in] max Max clip value as attribute
+    * @return std::shared_ptr<Node> 
+    */
+    std::shared_ptr<Aidge::Node> Clip(
+        const std::string &name = "",
+        float min = std::numeric_limits<float>::lowest(),
+        float max = std::numeric_limits<float>::max()
+        );
+
+}
+namespace {
+template <>
+const char* const EnumStrings<Aidge::ClipAttr>::data[]
+    = {"min", "max"};
+}
+
+#endif /* AIDGE_CORE_OPERATOR_CLIP_H_ */