Duals all NaN when including ReserveDemandCurve

[Nongchao]

When including PSY.ReserveDemandCurve in ProblemTemplate, the problem solves, but the duals are all NaN.

[jd-lara]

@sourabhdalvi Do you call what the implementation of ReserveDemandCurve looks like?

[Nongchao]

@sourabhdalvi Do you call what the implementation of ReserveDemandCurve looks like?

Sourabh helped me figured out what's causing the issue. Essentially SOS variables were introduced and my understanding is there's no way to fix these variables to solve a LP to get duals. But in my application, SOS variables are not needed, so we had a "hacky" way to by-pass the convexity check and not introducing SOS variables for now.

[jd-lara]

@Nongchao I'd be good to know what you are doing to make sure that we can incorporate the fix you are using.

[jd-lara]

closing since it can't be replicated. @rodrigomha we need a test with the new ORCD to recover Duals