Merge branch 'dev' into 'sub-backward'

# Conflicts:
#   unit_tests/operator/Test_MetaOperator.cpp

include/aidge/backend/cpu/operator/AddImpl.hpp

@@ -33,8 +33,6 @@ using AddImpl_cpu = OperatorImpl_cpu<Add_Op,
          const std::vector<std::size_t>&, 
          const std::vector<std::size_t>&, 
          const void*, 
-         const void*, 
-         const void*, 
          void*, 
          void*)
 >;

include/aidge/backend/cpu/operator/AddImpl_kernels.hpp

@@ -154,15 +154,11 @@ void AddImpl_cpu_backward_kernel(const std::size_t input0Length,
                                const std::vector<std::size_t>& dims0,
                                const std::vector<std::size_t>& dims1,
                                const std::vector<std::size_t>& outputDims,
-                               const void* input0_,
-                               const void* input1_,
                                const void* grad_output_,
                                void* gradientInput0_,
                                void* gradientInput1_)
 {
     // TODO: Remove input0/1 from the function
-    const I* input0 = static_cast<const I*>(input0_);
-    const I* input1 = static_cast<const I*>(input1_);
     const O* gradOutput = static_cast<const O*>(grad_output_);
     auto* gradInput0 = static_cast<I*>(gradientInput0_);
     auto* gradInput1 = static_cast<I*>(gradientInput1_);

src/operator/AddImpl.cpp

@@ -73,8 +73,6 @@ void Aidge::AddImpl_cpu::backward() {
                in0->dims(),
                in1->dims(),
                out0grad->dims(),
-               getCPUPtr(in0),
-               getCPUPtr(in1),
                getCPUPtr(out0grad),
                getCPUPtr(in0grad),
                getCPUPtr(in1grad));