Update to CODATA 2019

[bmsprague]

Constants should be updated to CODATA 2019 values, as follows:

Universal gas constant, J/(kmol·K) = 8314.46261815324D0 ! CODATA 2019 exact value (Nₐ × k)
Avogadro constant × 10⁻²³ = 6.02214076D0 ! CODATA 2019 exact: 6.02214076 × 10²³ mol⁻¹
Boltzmann constant × 10²³ = 1.380649D0 ! CODATA 2019 exact: 1.380649 × 10⁻²³ J/K

As thermo.lib provides h/R, s/R, no scaling of thermo data is necessary.

[markleader]

Thanks for the suggestion. At this stage, the constants are intentionally kept consistent with CEA2 to support validation and regression testing against the legacy code. Maintaining numerical equivalence during this initial public release is a priority.
That said, updating to CODATA values is a reasonable future change once the validation baseline is fully established. I’ll leave this issue open to track that potential update and revisit it when we define a post-CEA2-compatibility roadmap.

[jphill4]

I was having a similar discussion about another NASA physics library and we ultimately concluded that this should be part of the input dataset, not just hard-coded in the parameters module. That way we can swap back and forth without rebuilding the code and maintain direct traceability to legacy test cases. I would propose adding another optional keyword to the problem specification in the input file that allows selecting the CODATA standard to use and/or directly specifying values for individual constants. That should be do-able in a backwards-compatible manner.