Harmonize / standardize variable names

[dlebauer]

Summary

There are many duplicate variable names, and they don't follow the same conventions.

• Duplicates should be merged
  • in some cases, this will require transforming values in the traitsandyieldsview table
  • both means and stats should be transformed (only stats that are on the same scale as mean, e.g. SE, SD; not p or CV)
• Each record should have a standard_name
• (optional) names should be more standard

This spreadsheet includes

• proposed standard names (we can accept these as proposed)
• duplicates identified