Figure out PUA characters that look like accents

[rimas-kudelis]

There are also interesting glyphs assigned to positions U+EFB0-U+EFB5 and U+F004-U+F005. I'll probably have to ask Fotonija people about what they mean and what they were used for.

[rimas-kudelis]

I think I have it figured out with the help of some googling and a book that I have:

• book about Palemonas font
• How to Find the Character You Need in New Athena Unicode Font (v.5)
• Metrical Notation: A Guide to the most frequently-used symbols in Unicode

Glyph	Notes
U+EFB0	Palemonas (page 33): U+E0C7 DAUGISKAITOS SU UMLIAUTU ŽENKLAS Approx. translation: PLURAL WITH UMLAUT SIGN New Athena Unicode (page 11): U+EC3C or U+2012 U+0308 contracted biceps = two dots over macron
U+EFB1	Palemonas (page 36): U+E0CC KIRČIUOTAS ILGASIS SKIEMUO Approx. translation: STRESSED METRICAL LONG
U+EFB2	Potentially same as: U+23D3 METRICAL SHORT OVER LONG
U+EFB3	Palemonas (page 33): U+E0C8 BRŪKŠNYS SU TAŠKU Approx. translation: MINUS DOT Listed next to U+2238 DOT MINUS, and looks like its mirrored sibling.
U+EFB4	Palemonas (page 36): E0C9 KIRČIUOTA MORA Approx. translation: STRESSED METRICAL BREVE
U+EFB5	Identical to U+21E7 UPWARDS WHITE ARROW

I think I'm going to remove the characters above.

• In case of U+EFB0, it's still unclear to me whether the sign was supposed to be linguistic or mathematical. Plus, both Palemonas and NAU have the two dots much closer to the line, so it's potentially not even supposed to be the same character.
• In case of U+EFB1, U+EFB2, and U+EFB4, these characters don't compliment an existing set, because Jonova has no METRICAL characters that are part of Unicode, so providing only a few supplemental nonstandard ones doesn't seem logical to me.
• U+EFB3 complements an existing character but it's unclear why this complement has no official Unicode assignment if it's actually useful.
• U+EFB5 is just a duplicate of U+21E7. I see no reason to keep it.

As for U+F004 and U+F005, these are actually used as components of Latvian letters with cedillas. The glyphs likely should stay, but I'm not so sure regarding their code assignments – I suspect they might be just an artifact of some limitations back in the day. I've already renamed these glyphs to sign.cedilla.commabelow and sign.cedilla.commaturnedabove and should probably turn them into alt forms of U+00B8 CEDILLA and free up the PUA codepoints.