[Bug] Broken import of referencePorosity from external meshes

[klevzoff]

Describe the bug
Any input deck attempting to import referencePorosity doesn't work after #1311 .
I haven't checked, but suspect that VTK output of referencePorosity will be invalid for the same reason.

To Reproduce
Steps to reproduce the behavior:

1. Run src/coreComponents/physicsSolvers/fluidFlow/benchmarks/SPE10/dead_oil_spe10_layers_83_84_85.xml
2. See error

Importing field data from mesh dataset
***** ERROR
***** LOCATION: /home/klevtsov/work/GEOSX-develop/src/coreComponents/common/TypeDispatch.hpp:280
***** Controlling expression (should be false): true
***** Rank 0: Type LvArray::Array<double, 1, camp::int_seq<long, 0l>, long, LvArray::ChaiBuffer> was not dispatched.
Check the stack trace below and revise the type list passed to dispatch().
If you are unsure about this error, please report it to GEOSX issue tracker.

** StackTrace of 5 frames **
Frame 0: geosx::ProblemManager::importFields() 
Frame 1: geosx::ProblemManager::problemSetup() 
Frame 2: geosx::GeosxState::initializeDataRepository() 
Frame 3: main 
Frame 4: __libc_start_main 
Frame 5: _start 
=====

Expected behavior
Should run.

Platform (please complete the following information):

• Machine: any
• Compiler: any
• GEOSX Version: develop (as of a23824a6)

Additional context
The problem is due to the fact that external field import assumes every field coming from a material has a quadrature dimension, which holds for all fields except referencePorosity. A simple fix is to make referencePorosity a 2D array but make its second dimension equal to 1. Otherwise, we need to have a conversation about how physics-agnostic code (input/output mainly) is supposed to handle a variety of fields some of which may or may not have the extra dimension.