Factor/To do/Unicode

• Normalize output streams
• Implement line and sentence breaks
• Unicode line breaks for string-lines and stream-readln
• BIDI
• CLDR--what do we need from here?
• Unicode number input
• Collation tailoring, break tailoring
• A convenient interface to all Unicode properties in the UCD
• Some support for Unihan information
• Performance/cleanup/better data structures
• Figure out how to stop growing image size with Unicode stuff

Done:

• Fix normalization errors
• Fix grapheme break errors
• Implement word breaks
• Write docs