Add IRR to cashloan and migrate functions to lib_financial.h

[cpaulgilman]

Fix #638

1. Add IRR output to cashloan

• IRR output available for Residential and Commercial financial models.
• Add nominal discount rate output to help with identifying nominal discount rate that results in NPV = 0 (useful for parametrics on real discount rate)
• Note that default Residential and Commercial cases use 100% debt, so IRR is NaN for default configurations because initial investment is zero
• Not modifying SAM UI to add IRR to Metrics table on Summary tab, so it is only available in Data Tables because IRR metric is a bit ambiguous for projects that do not earn revenue

2. Move financial model functions to lib_financial for all financial models

Task	SO	CS	TPO HD	MP	EPF	LPF	SLB	TPO H	CashLoan
Compare functions to SO before moving	NA	☑	☑	☑	☑	☑	☑	NA	NA
Check depreciation_sched_custom()	☑	☑	☑	☑	☑	☑	☑	NA	NA
Delete using namespace libfin from cmod	☑	☑	☑	☑	☑	☑	☑	☑	☑
Delete libfin functions from cmod	☑	☑	☑	☑	☑	☑	☑	☑	☑
Convert npv()to libfin::	☑	☑	☑	☑	☑	☑	☑	☑	☑
Convert irr()to libfin::	☑	☑	☑	☑	☑	☑	☑	☑	☑
Convert min()to libfin::	☑	☑	☑	☑	☑	☑	☑	NA	☑
Convert max()to libfin::	☑	☑	☑	☑	☑	☑	☑	NA	☑
Convert ppmt()to libfin::	☑	☑	☑	☑	NA	☑	☑	NA	☑
Convert payback() to libfin::	NA	NA	NA	NA	NA	NA	NA	*	☑

* Note that Third Party Ownership - Host model has payback cashflows but does not use the payback() function to calculate payback period.

Question

That there are special min() and max() functions defined in lib_financial. I changed all calls to min() and max() in the financial cmods to libfin::min() and libfin::max() to avoid conflict, but should any of those be std::min() or std::max() instead?

[sjanzou]

For Mac and Linux GitHub Actions to pass, please add the following to the lib_financial.cpp file for DBL_MAX to be defined:

DBL_MAX is defined in float.h and was previously included at the top of each financial compute module as
#ifndef WIN32
#include <float.h>
#endif

[sjanzou]

DBL_MAX was not defined and is causing the Linux and MacOS tests to fail.

DBL_MAX is defined in float.h and was previously included at the top of each financial compute module as
#ifndef WIN32
#include <float.h>
#endif