visual expressiveness and perceptual discriminability metrics don't that into account symbols with multiple graphicalElements defining them

[rBoncorps]

When extracting concrete symbols, if there are multiple graphicalElements in a composite that defines a symbol, they are putten in a list of visual representation. When calculating visual expressiveness and perceptual discriminabilty metrics, we only that into accoount symbols that have only one visual representation (1 graphical element). That's because, it is not possible to determinate the shape, size, position and colour of the resulting aggregation of those visual representation.