Fix line Hessian bug

[nkoukpaizan]

Merge request type

• New feature
• Resolves bug
• Documentation
• Other

Relates to

• OPFLOW
• SOPFLOW
• SCOPFLOW
• TCOPFLOW
• CMake build system
• Spack configuration
• Manual
• Web docs
• Other

This MR updates

• Header files
• Source code
• CMake build system
• Spack configuration
• Web docs
• Manual
• Other

Summary

This fixes a bug where Bft was used in line Hessians, instead on Btf. The implementation now matches the comment above it in pbol.cpp.

• This has no effect on whether the current tests pass or fail. Btf=Bft when the phase/angle or tap ratio are 0, which is the case for case9, case118 and case_ACTIVSg200.
• All models had the same bug, so comparing one to the other cannot identify the bug. This is an additional argument in favor of unit tests that compare to analytical solutions.
• The Hessians are symmetric, and this seems to only affect the lower triangle. So it will have no effect if the solvers only use the upper triangular.