fix(semantic): preserve abstracts after list markers

[ehz0ah]

Description

Fix generated abstract truncation so numbered-list markers such as 1. are not mistaken for sentence endings.

A long overview introduction beginning with OpenViking's own numbered format could previously truncate .abstract.md to exactly 1., silently completing semantic processing with an unusable summary. The new boundary scanner ignores numbered/lettered list markers, decimals, and common abbreviations while preserving CJK punctuation, paragraph boundaries, Markdown structure, tiny limits, and the existing long-first-sentence behavior.

Human Involvement

• A human participated in the implementation or review loop
• This PR was generated entirely by AI agents without human participation in the loop

Related Issue

Fixes #3709

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactoring (no functional changes)
• Performance improvement
• Test update

Changes Made

• Detect contextual sentence and paragraph boundaries instead of treating every period-before-space as terminal.
• Ignore numbered/hierarchical/lettered list markers, decimals, and conservative common abbreviations.
• Add regression coverage through _normalize_overview_generation and the default 256-character abstract limit.

Testing

• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• I have tested this on the following platforms:
  • Linux
  • macOS
  • Windows

pytest tests/storage/test_semantic_processor_l0_l1.py
# 27 passed

broader semantic-processor suite
# 108 passed

ruff check openviking/storage/queuefs/semantic_processor.py tests/storage/test_semantic_processor_l0_l1.py
ruff format --check openviking/storage/queuefs/semantic_processor.py
git diff --check
# passed

The complete semantic-processor glob has one unrelated existing failure in the move/vector-store test caused by an unchanged path-lock mock.

Checklist

• My code follows the project's coding style
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Screenshots (if applicable)

Not applicable.

Additional Notes

This PR intentionally does not change abstract_max_chars or overview_max_chars defaults.