feat(sfr): add support for CELLIDs with zero values for unconnected reaches

[jdhughes-usgs]

• add deprecation waring for use of NONE for unconnected reaches
• update mf6io with SFR changes
• update ReleaseNotes with SFR changes
• add test of new SFR functionality for unconnected reaches

[langevin-usgs]

I'm surprised that this worked, because in a preceding call a couple of lines up from the SFR modification there is this:

this%igwfnode(n) = this%dis%noder_from_cellid(cellid, this%inunit, &
                                                      this%iout, &
                                                      flag_string=.true.)

In looking at noder_from_cellid and the call to nodeu_from_cellid it would seem that we should have to specify allowzero to be .true. Might be worth running that through the debugger to see why/how it works.

[mwtoews]

There is a potential issue with

modflow6/doc/mf6io/mf6ivar/examples/gwf-sfr-example.dat

Line 32 in 049db53

16 1 510 2500. 10 1.81E-03 1077.727 3.0 0.00003 0.03 1 1.0 0

which should be changed to something like:

- 16 1  510 2500.  10    1.81E-03 1077.727    3.0 0.00003  0.03     1   1.0    0 
+ 16 1 5 10 2500.  10    1.81E-03 1077.727    3.0 0.00003  0.03     1   1.0    0 

are these example files actually tested in any way?

[langevin-usgs]

There is a potential issue with

modflow6/doc/mf6io/mf6ivar/examples/gwf-sfr-example.dat

Line 32 in 049db53

16 1 510 2500. 10 1.81E-03 1077.727 3.0 0.00003 0.03 1 1.0 0

which should be changed to something like:

- 16 1  510 2500.  10    1.81E-03 1077.727    3.0 0.00003  0.03     1   1.0    0 
+ 16 1 5 10 2500.  10    1.81E-03 1077.727    3.0 0.00003  0.03     1   1.0    0 

are these example files actually tested in any way?

Good find, @mwtoews. These files are not used in any way other than being written as an example in the mf6io guide.