fix(UzfCellGroup): SQUARE_GWET option records ET as inflow instead of outflow

[emorway-usgs]

GWET values calculated using the SQUARE_GWET option in the UZF options block had the wrong sign.

• Replaced section above with description of pull request
• Resolves discussion UZF-GWET term on volume budget #1904
• Adds a new test
• Ran ruff on new and modified python scripts in autotests subdirectory.
• Formatted new and modified Fortran source files with fprettify
• Updated develop.tex with a plain-language description of the bug fix, change, feature; required for changes that may affect users

[emorway-usgs]

Some further notes on GWET as calculated by UZF:

For the SQUARE_GWET option in UZF, thcof (shorthand for "total uzf hcof contribution to GWF model") in the etfunc_nlin is not a function of head except over the smoothing interval at the extinction depth. This nonlinearity is handled directly through the Jacobian in the uzf8 source file:

    ! -- store derivative for Newton addition to equations in _fn()
    this%deriv(i) = uzderiv + derivgwet

Also, note that in the call to etfunc_lin, thcof is updated and returned to simgwet. However, in the function etfunc_nlin, thcof is not updated and remains zero. As result, in the following code, found in simgwet

if (igwetflag == 1) then
  et = etfunc_lin(s, x, c, det, trhs, thcof, hgwf, &
                  this%celtop(icell), this%celbot(icell))
else if (igwetflag == 2) then
  et = etfunc_nlin(s, x, c, det, trhs, thcof, hgwf)
  signflip = -1
end if
! this%gwet(icell) = et * this%uzfarea(icell)
trhs = trhs * this%uzfarea(icell)
thcof = thcof * this%uzfarea(icell)
this%gwet(icell) = (trhs - (thcof * hgwf))

the last line will always be trhs (shorthand for "total uzf rhs contribution to GWF model") minus 0.0 since thcof will remain at its initialized value of DZERO for the SQUARE_GWET option. As a result, %gwet will be positive when calculated using the SQUARE_GWET option but will be negative when the LINEAR_GWET option is used.

[langevin-usgs]

Nice job digging this out. Couple of comments included in this review.

[langevin-usgs]

Is there any way to adjust etfunc_lin and etfunc_nlin so that they return hcof, rhs, and rate in a way that is consistent with one another? It's unfortunate that there has to be sign shenanigans on the outside of the calls to these routines. Just a thought.

[emorway-usgs]

Have shifted etfunc_lin and etfunc_nlin over to a dedicated file that is included in the unit testing.

[emorway-usgs]

Added a new test problem with checks to ensure the GWET is being removed and that the resulting heads are as expected. The test compares results when using the classic LINEAR_GWET vs SQUARE_GWET