[WIP] Records prototype

[ChrisPenner]

• Implement ABT node
• Implement ABT hashing
• Implement ABT serialization/deserialization
• Implement record term parser/printer
• Implement record type parser/printer
• Implement kind checking
• Implement type synthesis
• Implement type unification
• Implement runtime hashing
• Custom error for missing field types
• Custom error for mismatched field types
• Custom error for extra/additional field types
• Implement pattern match parsing
• Implement pattern match type-checking
• Implement ANF generation
• Implement runtime evaluation
• Implement record field accessor parsing
• Implement record field accessor typechecking
• Implement record field accessor runtime
• Implement decompilation
• Implement Universal eq/ord etc.
• Add Backreferences to original record types for better runtime crash messages
• Add more closure cases for faster packing/unpacking into small records
• Fix TDNR within record fields that doesn't work for some reason.

Testing

• Verify we can save record values in codebase
• Verify we can embed records in union types (edited)