Hydraulic Library Update

[bradcarman]

Implements the following changes:

• fixes regPow
• implements p(rho) in place of rho(p) to enable Rodas5P adaptive solving, see Rodas5P Goes unstable with an increase in abstol OrdinaryDiffEq.jl#2561
• implements Initialization System in components and tests

Note: a new version release is needed, DynamicVolume function has changed, removing the N input because discretization is not possible with the Initialization System and solving with Rodas5P. This was an advanced feature that can be added back in the proper way with a future research project.

[ChrisRackauckas]

The initial Jacobian is singular, the initialization solution has an index-2 DAE at the starting point of the time stepping.

Main.LinearAlgebra.cond(W) = 1.9507933081280378e12
Main.LinearAlgebra.cond(W) = 1.8757802497588914e12
Main.LinearAlgebra.cond(W) = 1.9100754482497002e12
Main.LinearAlgebra.cond(W) = 1.9291032293541746e12
Main.LinearAlgebra.cond(W) = 1.9449040001602686e12
Main.LinearAlgebra.cond(W) = 1.956880525752231e12
Main.LinearAlgebra.cond(W) = 1.8642642846288142e12
Main.LinearAlgebra.cond(W) = 1.8840317888731355e12
Main.LinearAlgebra.cond(W) = 1.896847421636734e12
Main.LinearAlgebra.cond(W) = 1.925558550278289e12
Main.LinearAlgebra.cond(W) = 1.8469820435050667e12
Main.LinearAlgebra.cond(W) = 1.873127333970352e12
Main.LinearAlgebra.cond(W) = 1.85940822577763e12
Main.LinearAlgebra.cond(W) = 1.9505801274429348e12
Main.LinearAlgebra.cond(W) = 1.9439521531702615e12
Main.LinearAlgebra.cond(W) = 1.9630631453906282e12

[mgithubk]

@ChrisRackauckas and @bradcarman - it looks like the hydraulics setting allow_symbolic to true cleared the hydraulics errors that showed up in Chris' check.

There were questions about whether or not this method will scale though - making its use for a library questionable.

[bradcarman]

allow_symbolic is scalable. Presumably this option will be default to true in future ModelingToolkit releases, and therefore this PR is only an improvement on the current version.

[ChrisRackauckas]

@matthew-kapp Can you look at updating this one? Your change to macros is what's blocking this PR, so this needs to be merged with that change.

[matthew-kapp]

@ChrisRackauckas @bradcarman

I see the test failure's are unrelated to hydraulic - all hydraulic tests pass. Could it make sense to rebase first? Or does a rebase result in conflicts?

Also, on first inspection, the changes made are still using the functional form. I did NOT refactor the following models:

Is it possible to see specifically which models are creating conflicts? I can't view the conflicts on my side.