Add capacity to read-in gridded forcings when executed in standalone

[GreyREvenson]

Purpose

This PR adds capacity to read-in gridded forcings when executed in standalone (i.e., outside of NextGen).

Additions

• This PR adds two type-bound subroutines in ReadForcings and SetForcings. These subroutines were added to forcinggrid_type in ForcingGridType.f90.
• The ReadForcings subroutine finds/opens/reads the forcing file for an argument date. The forcings are read-into newly added 'read' arrays (e.g., forcingsgrid_type%read_pcprate holds read-in values for precipitate forcings).
• The SetForcings subroutine transfers forcing values from the added 'read' arrays to the forcinggrid_type arrays that hold forcings for the current time step (e.g., values in forcingsgrid_type%read_pcprate are transferred to forcingsgrid_type%PRCP).

Changes

• Before this PR, domaingrid_type%nowdate was calculated within a call to UtilitiesMain via solve_noahowp. domaingrid_type%nowdate was therefore recalculated for every grid cell. After this PR, domaingrid_type%nowdate is calculated with a call to advance_in_time (in RunModule.f90), which facilitates passing domaingrid_type%nowdate as an argument to the newly added ReadForcings and SetForcings subroutines and prevent the unnecessary recalculation of domaingrid_type%nowdate with every grid cell.

Testing

• Hourly AORC forcings were obtained for the North Fork (WY) test basin from Jason
• A py script was created to aggregate the hourly forcings into daily, monthly, and yearly forcing files
  • Please contact @GreyEvenson-NOAA for these forcings as they are too big to upload to GitHub
• The PR modified code was tested once for each of the forcing aggregation time steps (i.e., hourly, daily, monthly, yearly)

[GreyREvenson]

Tagging @SnowHydrology and @andywood is case you'd like to provide input/feedback on this PR's general approach.

@andywood previously wrote that 'In Fortran the ability to read large chunks at once is really valuable. The huge thing to avoid is unnecessary file open/close/inquire/read/write operations -- so we hopefully want to facilitate this.' This PR has the program reading in the whole of any given forcings file in one-go, which I believe takes advantage of Fortran's relative strength in reading in large arrays. However, the approach could be problematic if the user doesn't have the computing resources to read-in such a large file.

[drakest123]

This PR is an optimization that reduces the number of times that a call that advances the model simulation time is invoked.

• Compiled and executed: run/noah_owp_modular.exe
• test/noahowp_test.exe ../run/namelist.input > output_test
• No errors were detected. The results of the test output were not posted on Github for comparison with the output generated from this test.