bugfix for handling unstructured input data with multiple levels

[mvertens]

Description of changes

This fixes #25

Specific notes

This PR also does the following:

• new error checking for most allocations
• cleanup of stdout formatting
• replacement of character(*) with character(len=*)
• introduction logunit and mainproc in shr_strdata_type and both logunit and mainproc in shr_stream_streamType. This is needed since the inline interface was not always writing all log output consistently.
• new expanded API for setting stream pointers with optional arguments in dshr_strdata_mod.F90 for
  shr_strdata_get_stream_pointer_1d and shr_strdata_get_stream_pointer_2d

 subroutine shr_strdata_get_stream_pointer_1d(sdat, strm_fld, strm_ptr, rc, requirePointer, errmsg) 
    type(shr_strdata_type)     , intent(in)    :: sdat
    character(len=*)           , intent(in)    :: strm_fld
    real(r8)                   , pointer       :: strm_ptr(:)
    integer                    , intent(out)   :: rc
    logical,          optional , intent(in)    :: requirePointer
    character(len=*), optional , intent(in)    :: errmsg

and

 subroutine shr_strdata_get_stream_pointer_2d(sdat, strm_fld, strm_ptr, rc, requirePointer, errmsg) 
    type(shr_strdata_type)     , intent(in)    :: sdat
    character(len=*)           , intent(in)    :: strm_fld
    real(r8)                   , pointer       :: strm_ptr(:,:)
    integer                    , intent(out)   :: rc
    logical,          optional , intent(in)    :: requirePointer
    character(len=*), optional , intent(in)    :: errmsg

If requirePointer is not provided - then if the pointer is not round the subroutine returns without an error. If requirePointer is an argument and is true, than normally an errmsg is provided that with it describes why the pointer is required. Also - now if the pointer is required and is found - then the pointer is initialized to NaN.
The new setting of NaNs in the stream and state pointers resulted in the following additional changes that needed to be brought in:

• refactored dlnd code to remove presence of present nans in stream pointers
• fixed problem in drof that came up due to presence of presence of nans
• fixed problem in datm/cplhist that cam up due to presence of nans

Contributors other than yourself, if any: None

CDEPS Issues Fixed: #25

Are there dependencies on other component PRs: None

Are changes expected to change answers (bfb, different to roundoff, more substantial): bfb

Any User Interface Changes (namelist or namelist defaults changes): None

Testing performed:

• Correctly reading in nudging data at ne16pg3 with 58 vertical levels.
• aux_clm_noresm compared to ctsm5.4.002_noresm_v1 passed
• aux_blom_noresm compared to noresm3_0_beta09 passed
• aux_cam_noresm compared to noresm3_0_beta09 passed
• noresm_prealpha compared to noresm3_0_beta09 passed

Hashes used for testing:
noresm3_0_beta09 with this PR

[gold2718]

I found a few items that need fixing. Is there a way to test all the log output to make sure it is happening correctly?

[mvertens]

@gold2718 - thanks for your careful review - I think this is ready for another review.

[gold2718]

I still have some requests. Most of these are suggested improvements, however, there are a couple of statements which could cause problems or crash the run.

[mvertens]

@gold2718 - I've addressed all of your comments. Thanks for catching the problems.

[gold2718]

This looks okay now.