feat: Phase 2 complete — 13 Phases of COBOL type classification and test benchmark

P0.6: gcov infrastructure
P1: extract_structure output expansion (11 new feature fields)
P2: Confusion group rule engine (8 pairs + contradiction + backtrack)
P3: 4-factor confidence calculation + quality gate update
P4: 33+2 COBOL program type test samples (22 files, 7 categories)
P5: parametrized/ test data generation engine
P6: japanese_data.py lookup tables
P7-10: Type-specific test suites (~159 parametrized tests)
P11: Full classification pipeline (classify_program) + orchestrator integration
P12: Documentation (module-interfaces, test-plan v3.0, coverage-matrix)

Architecture decisions:
- classification_pipeline/ merged to hina/pipeline/
- parametrized/ as independent module
- japanese_data.py as root-level file
- hina/__all__ only exports classify_program()

Co-Authored-By: Claude <noreply@anthropic.com>

hina/__init__.py

@@ -1 +1,25 @@
-# HINA 程序分类与质量门禁包
+"""HINA 程序分类与质量门禁包
+
+公开 API:
+  classify_program()    — 完整类型判定管道（唯一外部入口）
+
+内部模块（不直接导出，但保留模块级导入以维持向后兼容）:
+  gate_check()          — 质量门禁判定
+  get_strategy()        — 策略模板获取
+  supplement()          — 策略补充
+  RetryHandler          — 分层重试处理器
+  collect_gcov()        — gcov 覆盖率采集
+"""
+
+from __future__ import annotations
+
+from .pipeline.pipeline import classify_program
+from .gate import check as gate_check
+from .strategy import get_strategy, supplement, supplement_only
+from .retry import RetryHandler
+from .gcov_collector import collect_gcov
+
+__all__ = [
+    # ═══ 唯一外部入口 ═══
+    "classify_program",  # (source: str, llm?: object) -> dict
+]