InterpretationEngine v0.1.0

[brownag]

This is draft PR for a v0.1.0 release of InterpretationEngine. With these additions it will be at least theoretically possible to fully evaluate primary rules provided input property data are available.

• initRuleset() now links hedge and operator functions to nodes of the rule set
• CVIRLinear() fix for certain specifications of linear function, update example
• .NULL_NOT_RATED() hedge function uses NaN to represent "Not rated"
  • Performing logic on mix of character and numeric is unwieldy; NaN allows data to be numeric but for "allowable" missing data to be distinguished from rules that should not generate a rating because required missing data are not present or some not rated condition is triggered
• Add interpret() method for a data.tree ruleset using either data.frame or SpatRaster input for property data
  • The SpatRaster method should be suitable for rasters that do not fit in memory, parallel processing, etc. though more testing is needed with more demanding use cases
• Miscellaneous documentation cleanup, several new packages in imports
• Remove message about experimental support for CrispExpression--this logic was added a couple years ago, has been tested, and works well

TODO:

• 0 and 1 row input to interpret() propdata needs to be handled

[brownag]

Plan is to merge this and continue refining documentation, examples, vignettes in another PR for actual v0.1 release. Everything changed in this PR is backward compatible based on my testing.