[Tensor] Fix Tensor::getIdx() method

The getIdx() method generates segfault on 0-rank coords
due to an out-of-bound access.
Also the computation was wrong for rank of coords
lower than rank of tensor by 2 or more.

include/aidge/data/Tensor.hpp

@@ -639,6 +639,7 @@ public:
      * the remaining coordinates are assumed to be 0.
      * Beware: the contiguous index will only correspond to the storage index
      * if the tensor is contiguous!
+     * Note that the coordIdx may be an empty vector.
      *
      * @param coordIdx Coordinate to an element in the tensor
      * @return DimSize_t Contiguous index
@@ -646,12 +647,13 @@ public:
     std::size_t getIdx(const std::vector<std::size_t>& coordIdx) const {
         AIDGE_ASSERT(coordIdx.size() <= mDims.size(), "Coordinates does not match number of dimensions");
         std::size_t flatIdx = 0;
-        std::size_t i = 0;
-        for(; i < coordIdx.size() - 1; ++i) {
-            AIDGE_ASSERT(coordIdx[i] < mDims[i], "Coordinates dimensions does not fit the dimensions of the tensor");
-            flatIdx = (flatIdx + coordIdx[i]) * mDims[i + 1];
+        for(std::size_t i = 0; i < mDims.size(); ++i) {
+            auto coord = i < coordIdx.size() ? coordIdx[i]: 0;
+            AIDGE_ASSERT(coord < mDims[i], "Coordinates dimensions does not fit the dimensions of the tensor");
+            auto nextDimSize  = i + 1 < mDims.size() ? mDims[i + 1]: 1;
+            flatIdx = (flatIdx + coord) * nextDimSize;
         }
-        return flatIdx + coordIdx[i];
+        return flatIdx;
     }
 
     /**