Configuration file and parameter read-in changes

[madMatchstick]

There are 4 main (mostly minor) config/param changes here:

1. Soil parameters exponent_primary and exponent_secondary are now optional params read-in from *_config.txt; defaults to 1.0 otherwise.
2. Both gw_storage and soil_storage no longer allow option to read-in as percentage %, but only literal value.
3. Adjusted all parameters in *_config.txt to match ranges from NWM
4. Adding [units] to *_config.txt

Additions

• Parameters soil_params.expon and soil_params.expon_secondary added to all (7) cat_*_bmi_config*.txt config files
• Include [units] for necessary params in all (7) cat_*_bmi_config*.txt.

Changes

1. Soils exponent_primary and exponent_secondary read-in via read_init_config_cfe(). Otherwise, set default value to 1.0
2. Parameters gw_storage and soil_storage must only be literal values in config
   • Both values adjust (computed from *_max and % char removed) in all (7) cat_*_bmi_config*.txt
   • Adjust read-in logic for literal value only (e.g. here)
   • Negative storage check is done in init_soil_reservoir()
   • Function decs & defs, read_init_config_cfe() & init_soil_reservoir(), change slightly as *_is_ratio,max_storage, etc. not being used
3. Cgw and max_gw_storage values were significantly too large, adjusted to comply with NWM ranges for all (7) cat_*_bmi_config*.txt.
4. read_init_config_cfe() checks for [units]
   • e.g. soil_params.depth=2.0[m] or if unitless soil_params.b=4.05[] or similarly, just soil_params.b=4.05
   • Only for unit-worthy params, initial read-in checks for existence of [units] in config, and prints warning if missing. E.g. see soil_params.satdk

Removals

• All instances of ratio/literal "tracker" variables
• init_reservoir_storage() function obsolete with literal values.
• soil_params.mult for all config.txt

Testing

1. make_and_run_bmi.sh results the same before and after changes.
2. BMI unit tests pass

Todos

• Remove // old code commented out once PR approved
• Update user docs/readmes/install for parameter adjustments

[lcunha0118]

I tested and it all look good. For future reference, we might want to include warnings when the parameter values are unrealistic (negative).