Miscellaneous potential bugs

[willpearse]

• The site IDs in the Baldridge dataset are non-consequential. Some values are being skipped (i.e. 689, 691, etc.). It may need to be dealt with ultimately. (@bweedop Baldridge Dataset Site IDs #10)
• anderson.2011 points vs. polygons problems. @willpearse says to maybe email Peter (anderson.2011 #15 )
• Convert from read.xls to read_xls. Convert read.xls to use read_xls from tidyverse #16
• Careful with NA checking in .df.melt and .matrix.melt as checking on all and any means that NAs cause errors. Drop and instead issue a warning? Not very user-friendly... (NA handling in .df.melt and .matrix.melt #9 @willpearse )