Well-posedness and scalarization in vector optimization