This is not a bug.

The hanger design cases (the HGR cases) are used to design the springs, and the data from these (fictitious) cases is what you see in the hanger reports. In most systems the data in the hanger report does not exactly match the results from your real load cases, because the springs don't exist in the hanger design cases.

A spring at best can only balance the system in one position. Any alteration from the load (or system stiffness) used to design the spring will yield results different from what you see in the hanger reports. The system stiffness will be different because the HGR cases don't have the spring, while your real (OPE, SUS, etc) cases do.