LFRic: (Closes #2189) fix bugs with missing maps in enter data directive

[arporter]

No description provided.

[codecov]

Codecov Report

All modified lines are covered by tests ✅

Comparison is base (f78be5c) 99.84% compared to head (56ad083) 99.84%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2199      +/-   ##
==========================================
- Coverage   99.84%   99.84%   -0.01%     
==========================================
  Files         343      343              
  Lines       45929    45926       -3     
==========================================
- Hits        45858    45855       -3     
  Misses         71       71              

Files	Coverage Δ
...rc/psyclone/domain/lfric/kern_call_acc_arg_list.py	100.00% <100.00%> (ø)
src/psyclone/domain/lfric/kern_call_arg_list.py	100.00% <100.00%> (ø)
src/psyclone/dynamo0p3.py	99.81% <100.00%> (+<0.01%)	⬆️
src/psyclone/tests/dependency_test.py	99.51% <100.00%> (-0.06%)	⬇️
src/psyclone/tests/dynamo0p3_basis_test.py	99.02% <ø> (-0.01%)	⬇️
src/psyclone/tests/dynamo0p3_multigrid_test.py	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[arporter]

Although this fixes the bug, it does not address the missing arrays used to look-up loop limits. This doesn't actually seem to be a problem as presumably they're not part of the kernel launch. However, I'm wondering whether it would be better to do this as part of #2157 which is redoing this whole thing properly.
EDIT: have looked in detail there's actually no overlap as things are still done differently in the LFRic API (the PSyIR dependence analysis is not used). Therefore, I'll keep this one separate.

[arporter]

This PR fixes a bug in the openacc 'enter data' generation for inter-grid LFRic kernels - the wrong cell map was being provided to the directive. Ready for review. Probably one for @rupertford as he's most familiar with this code but could also be @hiker, @sergisiso or @TeranIvy.

[rupertford]

Good stuff Andy, I have some questions and comments., most of which could come from my lack of understanding of the implementation.

[rupertford]

Just a few very minor issues now.

[arporter]

CI-permitting this is ready for another look. I've just fired-off the integration tests too.

[arporter]

@rupertford, just a nudge to say I could do with this when you're feeling up to doing a review :-)

[rupertford]

All outstanding issues have been addressed. CI is happy. Integration tests are happy. There is no performance degradation in integration tests. Approving.

[rupertford]

Waiting for new release until adding to master as Iva has completed her tests.