Add full decision writing pipeline: classify, extract, brainstorm, write, QA, export

New services (11 files):
- classifier.py: auto doc-type classification + party identification (Claude Haiku)
- claims_extractor.py: claim extraction from pleadings (Claude Sonnet + regex)
- references_extractor.py: plan/case-law/legislation detection (regex)
- brainstorm.py: direction generation with 2-3 options (Claude Sonnet)
- block_writer.py: 12-block decision writer (template + Claude Sonnet/Opus)
- docx_exporter.py: DOCX export with David font, RTL, headings
- qa_validator.py: 6 QA checks with export blocking on critical failure
- learning_loop.py: draft vs final comparison + lesson extraction
- metrics.py: KPIs dashboard per case and global
- audit.py: action audit log
- cli.py: standalone CLI with 11 commands

Updated pipeline: extract → classify → chunk → embed → store → extract_references
New MCP tools: 29 total (was 16)
New DB tables: audit_log, decisions CRUD, claims CRUD
Config: Infisical support, external service allowlist

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

mcp-server/src/legal_mcp/services/audit.py

@@ -0,0 +1,76 @@
+"""Audit log — תיעוד כל פעולה במערכת.
+
+כל פעולה מתועדת: מי, מתי, מה, על איזה תיק.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from datetime import datetime
+from uuid import UUID, uuid4
+
+from legal_mcp.services import db
+
+logger = logging.getLogger("audit")
+
+
+async def log_action(
+    action: str,
+    case_id: UUID | None = None,
+    document_id: UUID | None = None,
+    details: dict | None = None,
+    user: str = "system",
+) -> None:
+    """רישום פעולה ב-audit log.
+
+    Args:
+        action: סוג הפעולה (upload, classify, extract_claims, set_outcome, write_block, export, etc.)
+        case_id: מזהה תיק (אם רלוונטי)
+        document_id: מזהה מסמך (אם רלוונטי)
+        details: פרטים נוספים
+        user: מזהה המשתמש
+    """
+    pool = await db.get_pool()
+    async with pool.acquire() as conn:
+        await conn.execute(
+            """INSERT INTO audit_log (id, action, case_id, document_id, details, actor, created_at)
+               VALUES ($1, $2, $3, $4, $5, $6, $7)""",
+            uuid4(), action, case_id, document_id,
+            json.dumps(details or {}, ensure_ascii=False, default=str),
+            user, datetime.utcnow(),
+        )
+    logger.info("AUDIT: %s | case=%s | user=%s | %s", action, case_id, user,
+                json.dumps(details or {}, ensure_ascii=False)[:200])
+
+
+async def get_audit_log(
+    case_id: UUID | None = None,
+    action: str | None = None,
+    limit: int = 50,
+) -> list[dict]:
+    """שליפת audit log."""
+    pool = await db.get_pool()
+    conditions = []
+    params: list = []
+    idx = 1
+
+    if case_id:
+        conditions.append(f"case_id = ${idx}")
+        params.append(case_id)
+        idx += 1
+    if action:
+        conditions.append(f"action = ${idx}")
+        params.append(action)
+        idx += 1
+
+    where = f"WHERE {' AND '.join(conditions)}" if conditions else ""
+    params.append(limit)
+
+    async with pool.acquire() as conn:
+        rows = await conn.fetch(
+            f"SELECT * FROM audit_log {where} ORDER BY created_at DESC LIMIT ${idx}",
+            *params,
+        )
+
+    return [dict(r) for r in rows]