fix failed onnx tests

Context

This MR is intended to fix the failed onnx tests encountered in issue aidge_onnx#50 and issue aidge_onnx#46.

This includes the following problems:

• Flatten: missing python binding.
• And/Equal: A mistake was made, the operator Equal was named And.
• AvgPooling: unsupported attrs dilations and ceil_mode
• MaxPooling: unsupported attr dilations

Detailed major modifications

The following modifications were made:

• Add python binding for flatten
• Add Equal operator
• Add dilations attr to Maxpooling
• Add dilations and ceil_mode to AvgPooling

TODO

This MR would only fix the following:

• Add Flatten binding
• Add Equal Op
• AvgPooling: add dilations and ceil_mode attrs
• MaxPooling: add dilationsattr

include/aidge/operator/Equal.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_CORE_OPERATOR_EQUAL_H_
+#define AIDGE_CORE_OPERATOR_EQUAL_H_
+
+#include <memory>
+#include <string>
+#include <vector>
+
+#include "aidge/utils/Registrar.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/graph/Node.hpp"
+#include "aidge/utils/Types.h"
+
+namespace Aidge {
+
+/**
+ * @brief Tensor element-wise logical equal operation.
+ */
+class Equal_Op : public OperatorTensor,
+    public Registrable<Equal_Op, std::string, std::function<std::shared_ptr<OperatorImpl>(const Equal_Op&)>> {
+public:
+    static const std::string Type;
+
+    /**
+     * @brief Compute element-wise Equal operation on two given inputs.
+     * @details supports broadcasting of both operands.
+     */
+    Equal_Op() : OperatorTensor(Type, {InputCategory::Data, InputCategory::Data}, 1) {}
+
+    /**
+     * @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.
+     */
+    Equal_Op(const Equal_Op& op)
+        : OperatorTensor(op)
+    {
+        if (op.mImpl) {
+            SET_IMPL_MACRO(Equal_Op, *this, op.backend());
+        } else {
+            mImpl = nullptr;
+        }
+    }
+
+    /**
+     * @brief Clone the operator using its copy-constructor.
+     * @see Operator::Equal_Op
+     */
+    std::shared_ptr<Operator> clone() const override {
+        return std::make_shared<Equal_Op>(*this);
+    }
+
+    bool forwardDims(bool allowDataDependency = false) override final;
+
+    void setBackend(const std::string& name, DeviceIdx_t device = 0) override;
+    std::set<std::string> getAvailableBackends() const override;
+
+    static const std::vector<std::string> getInputsName(){
+        return {"data_input_1", "data_input_2"};
+    }
+    static const std::vector<std::string> getOutputsName(){
+        return {"data_output"};
+    }
+};
+
+inline std::shared_ptr<Node> Equal(const std::string& name = "") {
+    return std::make_shared<Node>(std::make_shared<Equal_Op>(), name);
+}
+} // namespace Aidge
+
+#endif /* AIDGE_CORE_OPERATOR_EQUAL_H_ */