Unicode 15 initial data files

[markusicu]

• Unihan data files from @Tsengtsz dec01
• new hardcoded CJK range for extension H
• initial UCD files from @Ken-Whistler dec02..03
• short & long block names in ShortBlockNames.txt, modified from Blocks.txt
• script codes Kawi+Nagm; run GenerateEnums for UcdPropertyValues.java
• generated files

Notes from Ken

The repertoire here covers ALL of the 15.0.0 additions, synched to the Pipeline page (and matching Michel's CDAM ballot draft as of 10/31, but not yet incorporating any CDAM ballot comment dispositions from later).

Note that there is one significant departure, to deal with the name collision for U+1DF27 LATIN SMALL LETTER N WITH LEFT HOOK. I've anticipated the most likely outcome and added "RAISED" into the names of 1DF25..1DF2A.

Notes from Ken about the initial drop for PropList.txt

This includes the non-automatic new property assignments:

1. Added Ideographic and Unified_Ideograph for Extension H (31350..323AF)

2. Added Other_Alphabetic for one Kannada mark (0CF3), one Khojki vowel
   sign (11241), and various Kawi signs and vowel signs (11F00.11F01,
   11F03, 11F34..11F3A, 11F3E..11F40).

3. Added Diacritic for 3 Arabic word signs (10EFD..10EFF) and for the
   Cyrillic modifier letters (1E030..1E06C).

4. Also added Other_Lowercase for the Cyrillic modifier letters.

5. Added Terminal_Punctuation and Sentence_Terminal for the two Kawi
   dandas (11F43..11F44), for general consistency with the way the danda
   and double danda are treated in related scripts. The rest of the Kawi
   punctuation is really murky, with no real analysis presented in the
   proposal, so I didn't make any assumptions that it would play in
   sentence break or even be terminal in position. (Kawi is one of the SE
   Asian scripts with no word spaces, so it ends up as lb=SA and requires
   special handling for paragraph formatting, anyway.)

Notes from Markus

2 new sets of decimal digits

11F50..11F59  ; Decimal # Nd  [10] KAWI DIGIT ZERO..KAWI DIGIT NINE
1E4F0..1E4F9  ; Decimal # Nd  [10] NAG MUNDARI DIGIT ZERO..NAG MUNDARI DIGIT NINE

[markusicu]

I regenerated the UCD files. No changes; in particular, no changes in DerivedNumericTypes/Values.

[markusicu]

TestInvariants fails. I sent an email discussing how to deal with incomplete data drops, such as Unihan data for new characters before even UnicodeData.txt has entries for new characters.