Get around the "deleted function" error

[einola]

Workaround for "deleted function" compilation error

Fixes #594

Task List

• [x] Defined the tests that specify a complete and functioning change (It may help to create a design specification & test specification)
• [x] Implemented the source code change that satisfies the tests
• [x] Documented the feature by providing worked example
• [x] Updated the README or other documentation
• [x] Completed the pre-Request checklist below

---

Change Description

When compiling with the Intel compiler, we get an error: the default constructor of "Nextsim::DynamicsKernel<6, 8>" cannot be referenced -- it is a deleted function. The solution proposed here is not to use the = default keyword but to manually define a constructor that sets the key parameters to sensible default values.

It is not clear why the other compilers don't flag this.

The compiler warns that warning #2196: routine is both "inline" and "noinline" for a few of the tests (and only tests), but I don't know how to fix that one.

---

Test Description

The model compiles and (almost) all tests run.

---

Documentation Impact

N/A

---

Other Details

N/A

---

Pre-Request Checklist

• [x] The requirements of this pull request are fully captured in an issue or design specification and are linked and summarised in the description of this PR
• [x] No new warnings are generated
• [x] The documentation has been updated (or an issue has been created to track the corresponding change)
• [x] Methods and Tests are commented such that they can be understood without having to obtain additional context
• [x] This PR/Issue is labelled as a bug/feature/enhancement/breaking change
• [x] File dates have been updated to reflect modification date
• [x] This change conforms to the conventions described in the README

Please complete your pull request description and delete the instructional text before submitting.

Congratulations and thank you for making your contribution to neXtSIM_DG!

[einola]

There are still two problems with this PR:

1. Not all the tests pass
2. The Intel compiler generates several warnings.

The TOPAZOn_test failure is:

===============================================================================
/cluster/home/einola/src/nextsimdg/physics/test/TOPAZOcn_test.cpp:28:
TEST SUITE: TOPAZOcean
TEST CASE:  TOPAZOcean test

/cluster/home/einola/src/nextsimdg/physics/test/TOPAZOcn_test.cpp:76: FATAL ERROR: REQUIRE( sst(45, 35) == -(0 + targetFrac) ) is NOT correct!
  values: REQUIRE( -0.035045 == -0.035045 )

The ConfigOutput_test failure is:

===============================================================================
/cluster/home/einola/src/nextsimdg/core/test/ConfigOutput_test.cpp:35:
TEST SUITE: ConfigOutput
TEST CASE:  Test periodic output

/cluster/home/einola/src/nextsimdg/core/test/ConfigOutput_test.cpp:35: FATAL ERROR: test case CRASHED: SIGSEGV - Segmentation violation signal

The ParaGrid_test failure is:

===============================================================================
/cluster/home/einola/src/nextsimdg/core/test/ParaGrid_test.cpp:457:
TEST SUITE: ParaGrid
TEST CASE:  Check if a file with the old dimension names can be read

/cluster/home/einola/src/nextsimdg/core/test/ParaGrid_test.cpp:457: ERROR: test case THREW exception: No such file or directory
file: ncFile.cpp  line:88: old_names.nc

The warning is warning #2196: routine is both "inline" and "noinline", and this is issued for:

• core/test/FileCallbackCloser_test.cpp(58)
• core/test/ModelArray_test.cpp(302)
• core/test/MonthlyCubicBSpline_test.cpp(57)
• core/test/Iterator_test.cpp(75)
• core/test/Time_test.cpp(241)
• physics/test/IceMinima_test.cpp(37)
• physics/test/SpecificHumidity_test.cpp(39)
• core/test/ModelArrayRef_test.cpp(202)
• dynamics/tests/DGModelArray_test.cpp(287,171)
• dynamics/tests/ParametricMesh_test.cpp(142)
• core/test/RectGrid_test.cpp(208)
• core/test/CommandLineParser_test.cpp(44)
• core/test/ConfiguredModule_test.cpp(129)
• core/test/Configurator_test.cpp(234)
• core/test/ConfigOutput_test.cpp(167)
• physics/test/ThermoIce0_test.cpp(164)
• core/test/ParaGrid_test.cpp(495)
• physics/test/ConstantOceanBoundary_test.cpp(36)
• core/test/ParaGrid_test.cpp(495)
• physics/test/BasicIceOceanFlux_test.cpp(148)
• core/test/ParaGrid_test.cpp(495)
• physics/test/SlabOcean_test.cpp(180)
• physics/test/IceGrowth_test.cpp(631)
• physics/test/BenchmarkBoundaries_test.cpp(83)
• physics/test/UniformOcean_test.cpp(93)
• physics/test/FiniteElementFluxes_test.cpp(304)
• core/test/PrognosticData_test.cpp(111)
• physics/test/ERA5Atm_test.cpp(118)
• core/test/ModelComponent_test.cpp(196)
• physics/test/TOPAZOcn_test.cpp(107)
• physics/test/ThermoWintonTemperature_test.cpp(360)

[einola]

I suggest we merge this and open a new issue to get the tests running using the Intel compiler.

[einola]

Unfortunately, neither the integration test, nor the dynamics benchmark run. In both cases, the model seg faults and dumps the core.