Add UAX #29 (Text Segmentation)

Implements UAX #29 for breaking text into grapheme clusters, words, and
sentences. Essential for text editors, search engines, and NLP applications.

Dependencies:
- UTS #51 (Unicode Emoji)

Key features:
- Grapheme cluster boundaries (user-perceived characters)
- Word boundaries for text selection and cursor movement
- Sentence boundaries for text processing
- Emoji ZWJ sequences and regional indicators
- Indic conjunct sequences for 10+ scripts
- 100% conformance (3,222/3,222 test cases passing)
- Unicode 17.0.0 support

Author: SCKelemen

README.md

@@ -184,6 +184,40 @@ for i := 1; i < len(breaks); i++ {
 
 [Full Documentation →](./uax14/README.md)
 
+### UAX #29: Text Segmentation (`uax29`)
+
+[![Go Reference](https://pkg.go.dev/badge/github.com/SCKelemen/unicode/uax29.svg)](https://pkg.go.dev/github.com/SCKelemen/unicode/uax29)
+
+Implementation of [UAX #29 (Unicode Text Segmentation)](https://www.unicode.org/reports/tr29/) for breaking text into grapheme clusters, words, and sentences.
+
+**Dependencies:** UTS #51 (Unicode Emoji)
+
+**Key Features:**
+- Grapheme cluster boundaries (user-perceived characters)
+- Word boundaries (text selection, cursor movement)
+- Sentence boundaries (text processing)
+- Emoji sequence handling with ZWJ
+- Regional indicator sequences (flag emojis)
+- Indic conjunct sequences
+- 100% conformance (3,222/3,222 test cases passing)
+
+**Quick Example:**
+```go
+import "github.com/SCKelemen/unicode/uax29"
+
+// Break text into grapheme clusters
+text := "👨‍👩‍👧‍👦 Hello"
+clusters := uax29.Graphemes(text)
+
+// Find word boundaries
+words := uax29.Words("Hello, world!")
+
+// Segment sentences
+sentences := uax29.Sentences("Hello Dr. Smith. How are you?")
+```
+
+[Full Documentation →](./uax29/README.md)
+
 ## References
 
 ### Metastandards

uax29/GraphemeBreakTest.txt

@@ -0,0 +1,115 @@
+# uax29 - Unicode Text Segmentation
+
+Implementation of [UAX #29: Unicode Text Segmentation](https://www.unicode.org/reports/tr29/) in Go.
+
+**Status:** Implemented with official Unicode test vectors (Unicode 17.0)
+
+## Overview
+
+This package will provide algorithms for breaking text into meaningful units:
+- **Grapheme clusters**: User-perceived characters (what users think of as "characters")
+- **Words**: Linguistic word boundaries for text selection and cursor movement
+- **Sentences**: Sentence boundaries for text processing
+
+## Planned Features
+
+### Grapheme Cluster Boundaries
+- Proper handling of combining marks (e.g., `e` + `́` = `é`)
+- Hangul syllable composition
+- Emoji sequences with Zero Width Joiner (ZWJ)
+- Regional indicator sequences (flag emojis)
+- Variation selectors
+
+### Word Boundaries
+- Alphabetic and numeric sequences
+- Proper handling of contractions (don't, can't)
+- Punctuation boundaries
+- CJK word segmentation (requires dictionary)
+- Hyphenated words
+
+### Sentence Boundaries
+- Period, question mark, exclamation handling
+- Abbreviation detection (Dr., Mrs., etc.)
+- Quote and parenthesis handling
+- Whitespace rules
+- Multiple punctuation handling (e.g., `...`, `?!`)
+
+## Use Cases
+
+- Text editors: cursor movement, selection, deletion
+- Search: tokenization and indexing
+- Natural language processing
+- Text-to-speech: proper phrase boundaries
+- Terminal UIs: text selection and wrapping
+
+## Implementation Status
+
+### Grapheme Cluster Boundaries ✅ (100.0% pass rate - 766/766) 🎉
+- **COMPLETE** implementation with Unicode 17.0 test vectors
+- Handles combining marks, Hangul syllables, all emoji sequences
+- Regional indicator pairs (flag emojis) working correctly
+- Prepend characters properly supported
+- Emoji modifiers (skin tones) correctly classified as Extend
+- GB11: Emoji ZWJ sequences fully implemented
+- GB9c: Indic conjunct sequences for 10+ scripts (Devanagari, Bengali, Gujarati, Oriya, Telugu, Malayalam, Myanmar, Balinese, Sundanese, Khmer)
+
+### Word Boundaries ✅ (100.0% pass rate - 1944/1944) 🎉
+- **COMPLETE** implementation with Unicode 17.0 test vectors
+- Handles all alphabetic/numeric sequences, contractions, punctuation
+- Regional indicator pairs with ZWJ transparency
+- Hebrew letter handling with single/double quotes
+- Katakana sequences and ExtendNumLet
+- Emoji sequences with modifiers and ZWJ
+- Proper handling of Format character exceptions
+
+### Sentence Boundaries ✅ (100.0% pass rate - 512/512) 🎉
+- **COMPLETE** implementation with Unicode 17.0 test vectors
+- Handles all sentence terminators (., ?, !, and many script-specific terminators)
+- Proper handling of abbreviations with ATerm
+- Complex Close* Sp* sequences correctly processed
+- SB8: Lowercase handling after ATerm Close* Sp*
+- SB8a: SContinue and sentence terminal sequences
+- SB9/SB10: Close and space handling after terminators
+- SB11: Breaking after sentence terminal sequences
+
+## Examples (Planned)
+
+```go
+// Grapheme clusters
+text := "👨‍👩‍👧‍👦"  // Family emoji (multiple codepoints)
+graphemes := uax29.Graphemes(text)
+// Returns 1 grapheme cluster
+
+// Words
+text := "Hello, world!"
+words := uax29.Words(text)
+// Returns: ["Hello", ",", " ", "world", "!"]
+
+// Sentences
+text := "Hello Dr. Smith. How are you?"
+sentences := uax29.Sentences(text)
+// Returns: ["Hello Dr. Smith. ", "How are you?"]
+```
+
+## Dependencies
+
+This package depends on:
+- **UTS #51 (Unicode Emoji)**: Provides authoritative emoji property data (Extended_Pictographic, Regional_Indicator, Emoji_Modifier, ZeroWidthJoiner constant)
+
+## Integration with Other Standards
+
+- **UTS #51 (Unicode Emoji)**: Emoji sequences are treated as single grapheme clusters per UAX #29 GB11
+- **UAX #14 (Line Breaking)**: UAX #29 word boundaries inform line break decisions
+- **UAX #9 (Bidirectional)**: Both needed for proper text layout
+- **UAX #11 (East Asian Width)**: Terminal cursor movement should respect grapheme boundaries
+
+## References
+
+- [UAX #29: Unicode Text Segmentation](https://www.unicode.org/reports/tr29/)
+- [Grapheme Cluster Boundaries](https://www.unicode.org/reports/tr29/#Grapheme_Cluster_Boundaries)
+- [Word Boundaries](https://www.unicode.org/reports/tr29/#Word_Boundaries)
+- [Sentence Boundaries](https://www.unicode.org/reports/tr29/#Sentence_Boundaries)
+
+## License
+
+MIT