Restore some properties to unicodetools

[macchiati]

1. Unicode JSPs used to support some properties that were not in the UCD, but useful. For example, a useful property is the ICU NFC_INERT, where NFC_INERT contains cp iff

For all strings X and Y, toNfc(X + cp + Y) = toNfc(X)+cp+toNfc(Y)

(That is far more useful than the Expands_On_NFC property that is in the UCD.

3. You used to be able to click on the two parts of the headers, such as in

Basic Latin — ASCII punctuation and symbols

and get to the characters that matched. That doesn't work any more. I think the problem is that

https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=%5B%3ABlock%3DBasic+Latin%3A%5D works =
https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=[:Block=Basic Latin:]

but the equivalent doesn't:
https://util.unicode.org/UnicodeJsps/list-unicodeset.jsp?a=\p{Block=Basic+Latin}

4. You get a message if an enum property is recognized but the value is invalid, eg
   [:block=nonvalue:]
   "The value 'nonvalue' is illegal. Values for Block must be in [Adlam, Aegean_Numbers,..."

At some point, probably a while now, you used to get a similar failure message if the property was not recognized. Now you just get null, eg
[:nonprop=nonvalue:]

5. The Z-Other category in https://util.unicode.org/UnicodeJsps/properties.jsp would be more useful if grouped accordingly (eg Emoji_DCM into the Emoji category), and contributory properties marked as such (maybe with the Source column, so Other_Alphabetic could have Source be UCD, Contributory). Might need new categories, eg for Do_Not_Emit_Preferred.

(The categories should be check against #18 to remove gratuitous differences)