Update UAX#29 text segmenter data rules to 16.0.

[makotokato]

Unicode 16.0.0 changes are property value only. So updating data files and test data.

[eggrobin]

Now that the issues with InCB in Unicode 15.1 have been resolved (see UTC-179-C26), we should get rid of our own derivation here:

icu4x/provider/source/src/segmenter/mod.rs

Lines 292 to 336 in 7e287f1

	// The Indic_Conjunct_Break property is separate from the Grapheme_Cluster_Break property.
	// See https://unicode.org/reports/tr44/#Indic_Conjunct_Break
	if p.name == "InCBConsonant" || p.name == "InCBLinker" || p.name == "InCBExtend"
	{
	let gcb_extend = gcb_name_to_enum
	.get_loose("Extend")
	.expect("property name should be valid!");
	for i in 0..(CODEPOINT_TABLE_LEN as u32) {
	if let Some(c) = char::from_u32(i) {
	let insc_value = insc.get(c);
	let sc = script.get(c);
	let is_gb9c_script = sc == Script::Bengali
	|| sc == Script::Devanagari
	|| sc == Script::Gujarati
	|| sc == Script::Malayalam
	|| sc == Script::Oriya
	|| sc == Script::Telugu;
	let is_incb_consonant = insc_value
	== IndicSyllabicCategory::Consonant
	&& is_gb9c_script;
	let is_incb_linker =
	insc_value == IndicSyllabicCategory::Virama && is_gb9c_script;
	// InCB = Linker or InCB = Consonant
	if (p.name == "InCBConsonant" && is_incb_consonant)
	|| (p.name == "InCBLinker" && is_incb_linker)
	// ZWJ is InCB=Extend, but is in a different GCB class anyway so
	// it needs to be special-cased in the tables.
	// NOTE(eggrobin): UAX #44, Version 15.1, instead excludes based
	// on InSC.
	// I believe that to be a defect in that version of Unicode.
	// This has been brought to the attention of the Properties and
	// Algorithms Group.
	|| (p.name == "InCBExtend"
	&& (gb.get32(i) == gcb_extend
	&& ccc.get32(i) != CanonicalCombiningClass::NotReordered
	&& !is_incb_consonant
	&& !is_incb_linker))
	{
	properties_map[c as usize] = property_index;
	}
	}
	}
	continue;
	}

Right now it is not causing any problems, since our custom derivation is consistent with what Unicode 16.0 does, but in 17.0 there will be substantial changes to the derivation, see UTC-179-C27 and https://www.unicode.org/reports/tr44/proposed.html#Derivation_InCB. These 17.0 changes do not come with any changes to the rules themselves, so the update should be just as easy as this one, but only if we are actually using the InCB from UTC.

[makotokato]

InCB isn't implemented in ICU4X yet, I will create a PR to add this property.

[aethanyc]

If updating the properties is all we need and no UAX29 rules needs a change, we should update these comments point to 16.0.0 and the new URLs.

icu4x/provider/source/data/segmenter/grapheme.toml

Lines 5 to 6 in b886706

	# These grapheme boundary rules are based on UAX #29, Unicode Version 15.1.0.
	# https://www.unicode.org/reports/tr29/tr29-43.html

icu4x/provider/source/data/segmenter/sentence.toml

Lines 5 to 6 in b886706

	# These sentence boundary rules are based on UAX #29, Unicode Version 15.1.0.
	# https://www.unicode.org/reports/tr29/tr29-43.html

icu4x/provider/source/data/segmenter/word.toml

Lines 5 to 6 in b886706

	# These word boundary rules are based on UAX #29, Unicode Version 15.1.0.
	# https://www.unicode.org/reports/tr29/tr29-43.html

[robertbastian]

nit: it's probably more efficient to use incb.iter_ranges_for_value than to call incb.get a million times

[Manishearth]

The code here does this already, I'm going to fix it in a followup

[Manishearth]

#6367