Code changes for files in the diag_manager directory

[MinsukJi-NOAA]

Description
Mixed mode changes for files in the diag_manager directory.

How Has This Been Tested?
CI unit tests all pass.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules
• New check tests, if applicable, are included
• make distcheck passes

[junwang-noaa]

I guess we need to implement like this if we have data array passed in with dimension.

[MinsukJi-NOAA]

Yes, I still don't understand the reason for this, but it allows the GNU unit tests to pass.

[junwang-noaa]

Maybe we already discussed this, would you please remind me why we need to comment out OVERLOAD_R8? My thinking is that 1) when we compile fms with r8 for mixed mode, we won't use "-DOVERLOAD_R8", so the code send*_r8 won't appear (just like you comment them off); 2) when we compile fms with r4 as before, we will use "-DOVERLOAD_R8", so the code can work as before. Is my understanding correct for 2)?

[MinsukJi-NOAA]

This is because send_data_2d and send_data_3d now use class(*) dummy arguments. This makes the interfaces of send_data_2d and send_data_2d_r8 indistinguishable, as I understand. Same with send_data_3d and send_data_3d_r8.

[binli2337]

Looks good.

[junwang-noaa]

@MinsukJi-NOAA Do you know why some CI test are cancelled?

[MinsukJi-NOAA]

The CI with parallel works is waiting for some computing resources. I suppose those may run successfully on GFDL's repository. But NOAA-EMC does not have parallel works resources associated with it. We should disregard those tests, and only look at Build libFMS test

[junwang-noaa]

I see, thanks.