Revise text in tuning the 2DAR selectivity

[iantaylor-NOAA]

This PR inspired by an emailed question and follows discussion in closed PR #241.

Changes are:

• add note about lack of generalized software for tuning 2DAR selectivity
• remove paragraph about stand-alone TMB model for tuning the rho parameters
• align terminology by using "2DAR" throughout the User Manual instead of mix of "2DAR" and "2DAR1".
• update Xu et al. 2020 reference to final published year

@HaikunXu, please feel free to look over this revised text in the SS3 User Manual and suggest any changes. The existing text was from 2017 and caused some confusion about the available software for tuning the selectivity parameters.

I've added the "2024" milestone to r4ss issue https://github.com/r4ss/r4ss/issues/500 to improve the tools available in r4ss for tuning 2DAR selectivity and if that effort is successful, we can revise this text again to reference whatever we come up with.

Thanks to @mdbryan for pointing out the need for revised text.

[e-perl-NOAA]

This looks good Ian. I removed TMB from the glossary but after those checks pass I will merge it in.

[HaikunXu]

I uploaded the stand-alone TMB + R code for the 2DAR option to GitHub (https://github.com/HaikunXu/2DAR4ss/tree/main). An user manual is provided to guide users on how to tune SigmaS and how to estimate the two autocorrelation coefficients outside SS.

[iantaylor-NOAA]

Thanks @HaikunXu! The SS3 user manual associated with the next release will include the links thanks to the addition by @e-perl-NOAA.