Document selectivity Pattern 21 & clean up inconsistent terminology in the selectivity descriptions

[iantaylor-NOAA]

I noticed when reviewing PR #236 that selectivity Pattern 21 has no documentation. This PR is attempt to fix that after looking through the source code and modifying the model at https://github.com/nmfs-ost/ss3-user-examples/tree/main/model_files/selex_length_example to add Pattern 21 in place of Pattern 24.

The modified files are in selex_length_example_3.30.22.1_type21.zip.

I made some changes to the documentation of the other selectivity patterns to align the text, including

• adding links between the descriptions of the three non-parametric length-based selectivity patterns (6, 21, and 43)
• "2*x" to "2*N breaks" or "2*N nodes" in the parameter counts
• "nages" and "Nages" to "N ages"
• "way point" to "break point"
• "set-up" to "setup"

This PR is based on the branch in #236, so I'm assuming that one gets merged first at which point this PR should only show changes to 9control.tex. I'm adding @e-perl-NOAA as reviewer to look over the hyperlinks and language in general, and @Rick-Methot-NOAA to make sure the description of Pattern 21 and small change to Pattern 6 are accurate.

[iantaylor-NOAA]

@Rick-Methot-NOAA, I just realized that I never finished sorting out this pull request from last month.

Could you check the new text starting at https://github.com/nmfs-ost/ss3-doc/pull/237/files#diff-6b9705ecd3c8837aa58aabbfea66ceaad0dd7aed3b75c386ac779d3802b7f197R2053 to confirm that description of selectivity type 21 is accurate? Most of it is drawn from my read of the code implementing that option, which hopefully was accurate, but some of the comments are more speculative, like "If the largest parameter is not fixed at 1.0 or estimated close to 1.0, then care should be taken in interpreting the associated fishing mortality values for the fleet in question."

Feel free to comment on the other changes, but that paragraph is the most significant one.