Currently individual subgraph compatibility results do not include information about target subgraph. This is generally fine as we generally should update single implementation detail at a time.... but in cases when we update few implementations in a PR we end up with multiple comments (one per implementation) and it is impossible to determine which result correspond to which implementation.