Clean up/finalize baseline LSTM

[lekoenig]

This issue captures several tasks that are needed to clean up/finalize the baseline LSTM:

• Add tmmn to list of meteorological drivers Rerun baseline LSTM with tmmn and tmmx #153
• Check date formatting for gridmet data
• Incorporate gridmet data in targets pipeline How to build/include .nc file #151
• Add additional segment/catchment attributes at NHDPlusv2-scale Gather additional static input variables for baseline LSTM v2 #51
• inputs_and_outputs_med_obs in 2a_model.R has redundant COMID columns
• Simplify pipeline by omitting targets for both well-observed and medium-observed sites Should 2a_model.R use med-obs or well-obs sites (but not both)? #148
• Decide whether provisional data will be included in the observations used for model training Use provisional NWIS data? #135
• Val times are currently defined in two places, and the values differ between those locations Decide whether to define val time period in _targets.R #160
• Flesh out README to describe the modeling inputs and steps, including how to modify and run the model Add model instructions to README #146