test: add 41 edge-case tests for chunking module

[voidborne-d]

Closes #430

Summary

Adds comprehensive edge-case test coverage for langextract/chunking.py, covering all untested code paths reported in #430.

New tests (41 total, organized into 11 focused test classes)

SentenceIteratorGuardTest (4 tests)

• Negative curr_token_pos → IndexError
• curr_token_pos past document end → IndexError
• curr_token_pos at exact end → StopIteration
• Mid-sentence start yields partial sentence first

CreateTokenIntervalTest (4 tests)

• Negative start index → ValueError
• Start equals end → ValueError
• Start greater than end → ValueError
• Valid interval construction

GetTokenIntervalTextTest (3 tests)

• start_index >= end_index → ValueError
• Equal start/end → ValueError
• TokenUtilError when tokenizer returns empty string for non-empty text

GetCharIntervalTest (2 tests)

• start_index >= end_index → ValueError
• Valid char interval with correct bounds

ChunkIteratorConstructorTest (6 tests)

• Both text and document None → ValueError
• text=None falls back to document.text
• Empty TokenizedText (no tokens) triggers re-tokenization
• Empty TokenizedText with document falls back to document.text
• String input is tokenized automatically
• No document → creates default Document

TextChunkErrorPathTest (2 tests)

• chunk_text raises ValueError without document
• char_interval raises ValueError without document

SanitizedChunkTextTest (3 tests)

• Whitespace normalization (newlines, consecutive spaces)
• Result is cached (same object on repeated access)
• _sanitize raises ValueError for whitespace-only input

LazyCachingTest (6 tests)

• _chunk_text starts as None, populated after first access, same object on repeat
• _char_interval starts as None, populated after first access, same object on repeat

MakeBatchesTest (4 tests)

• batch_size=1 → single-element batches
• batch_size=2 → two-element batches
• batch_size larger than total chunks → single batch
• Empty iterator → no batches

BrokenSentenceFlagTest (3 tests)

• Flag set to True when a sentence is split
• Flag reset to False after broken sentence fully consumed
• Flag stays False when all sentences fit

TextChunkStrTest + TextChunkPropertyCoverageTest (4 tests)

• __str__ shows "unavailable" when document is missing
• __str__ includes document ID and text when present
• document_id / document_text return None without document

Verification

$ python -m pytest tests/chunking_test.py tests/chunking_edge_cases_test.py -q
60 passed in 0.55s

All 41 new tests pass alongside the existing 19 tests.

[github-actions]

No linked issues found. Please link an issue in your pull request description or title.

Per our Contributing Guidelines, all PRs must:

• Reference an issue with one of:
  • Closing keywords: Fixes #123, Closes #123, Resolves #123 (auto-closes on merge in the same repository)
  • Reference keywords: Related to #123, Refs #123, Part of #123, See #123 (links without closing)
• The linked issue should have 5+ 👍 reactions from unique users (excluding bots and the PR author)
• Include discussion demonstrating the importance of the change

You can also use cross-repo references like owner/repo#123 or full URLs.