Updated MYNN surface layer submodule

[joeolson42]

This PR removes the old MYNN surface layer scheme and adds the new modularized submodule version. The scheme from the CCPP was used as the initial form but it has decomposed into 3 different modules for land, water, and ice. Some other changes (noted below) were made to accommodate this new version.

TYPE: bug fix, enhancement, new submodule version

KEYWORDS: MYNN-SFC, submodule

SOURCE: Joseph Olson (NOAA-GSL) and Xia Sun (NOAA-GSL)

DESCRIPTION OF CHANGES:
Problem:

• Different versions of this scheme exist in different models, making updates more tedious.
• Some minor bugs were found, such as SPP implementation to the thermal/moisture roughness lengths over water and reports of lacking restart reproducibility
• The use of shared namelist variables (iz0tlnd and isftcflx) carried different meaning for different surface layer schemes. Need MYNN-specific variables.

Solution:

• Centralize development in submodule repository and hook up to all models (WRF, MPAS, CCPP)
• implement bug fixes
• added new MYNN-specific options (sf_mynnsfc_sfcflux_land and sf_mynnsfc_sfcflux_water), which will be carried to all other model frameworks, and removed the use the shared options that carry different meaning in different surface layer schemes.

LIST OF MODIFIED FILES: list of changed files (use git diff --name-status master to get formatted list)
M .gitmodules
M Makefile
M Registry/Registry.EM_COMMON
M dyn_em/module_first_rk_step_part1.F
M main/depend.common
M phys/CMakeLists.txt
A phys/MYNN-SFC
M phys/Makefile
M phys/module_physics_init.F
D phys/module_sf_mynn.F
M phys/module_surface_driver.F

TESTS CONDUCTED:

1. Do mods fix problem? Yes.
2. All tests have been performed on the JET system with a recent version of the intel compiler. Tests were performed with a comprehensive set of debug flags. Additional testing is still in progress. Some minor updates are anticipated over the next couple weeks.
3. Are the Jenkins tests all passing?

RELEASE NOTE:
The MYNN surface layer scheme has been updated to the new modularized submodule version. The original form of this version was taken from the CCPP but it has been decomposed into 3 different modules for land, water, and ice. This helps to centralize all future development in submodule repository that will soon hook up to all models (WRF, MPAS, CCPP). Some minor bug fixes have been implemented, including a fix for using SPP in the water submodule. New MYNN-specific options (sf_mynnsfc_sfcflux_land and sf_mynnsfc_sfcflux_water) have been added to replace the shared options (iz0tlnd and isftcflx) that carried different meaning for different surface layer schemes.

[weiwangncar]

The regression test results:

Test Type              | Expected  | Received |  Failed
= = = = = = = = = = = = = = = = = = = = = = = =  = = = =
Number of Tests        : 23           24
Number of Builds       : 60           57
Number of Simulations  : 158           150        0
Number of Comparisons  : 95           86        0

Failed Simulations are: 
None
Which comparisons are not bit-for-bit: 
None

[joeolson42]

The latest regression tests (after the latest update):

Test Type | Expected | Received | Failed
= = = = = = = = = = = = = = = = = = = = = = = = = = = =
Number of Tests : 23 24
Number of Builds : 60 57
Number of Simulations : 158 150 0
Number of Comparisons : 95 86 0

Failed Simulations are: 
None
Which comparisons are not bit-for-bit: 
None