Fix claims parsing: truncated JSON recovery + chunking + compact output

config.py parse_llm_json: Added truncated JSON recovery. When Claude's
output is cut mid-JSON (common with long claim lists), the parser now:
- Finds the last complete JSON item (closing "}")
- Closes the array/object brackets
- Returns partial but valid results instead of None
Tested: recovers 2/3 items from truncated array, all cases pass.

claims_extractor.py:
- Prompt asks for compact output (150 words max per claim, group similar)
- Explicitly requests "no markdown, no explanations, JSON only"
- Long documents split into chunks at paragraph boundaries
- Each chunk processed separately, results merged
- max_tokens already at 8192

This fixes the recurring "0 claims" bug for committee responses and
permit applicant responses where the JSON was getting truncated.

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

mcp-server/src/legal_mcp/config.py

@@ -72,13 +72,19 @@ AUDIT_ENABLED = os.environ.get("AUDIT_ENABLED", "true").lower() == "true"
 # ── Utility ───────────────────────────────────────────────────────
 
 def parse_llm_json(raw: str):
-    """Parse JSON from LLM response, stripping markdown code blocks and extra text."""
+    """Parse JSON from LLM response, handling markdown wrapping and truncation.
+
+    Handles:
+    1. Markdown ```json ... ``` code blocks
+    2. Extra text before/after JSON
+    3. Truncated JSON (missing closing brackets) — attempts recovery
+    """
     import json
     import re
     raw = raw.strip()
     # Strip markdown code blocks
     raw = re.sub(r"^```(?:json)?\s*\n?", "", raw)
-    raw = re.sub(r"\n?\s*```$", "", raw)
+    raw = re.sub(r"\n?\s*```\s*$", "", raw)
     # Try direct parse first
     try:
         return json.loads(raw)
@@ -92,4 +98,38 @@ def parse_llm_json(raw: str):
                 return json.loads(match.group())
             except json.JSONDecodeError:
                 continue
+    # Attempt truncated JSON recovery:
+    # Find the start of JSON, then try closing open brackets
+    for opener, closer in [("[", "]"), ("{", "}")]:
+        start = raw.find(opener)
+        if start < 0:
+            continue
+        fragment = raw[start:]
+        # Try progressively removing trailing partial content and closing
+        # Look for the last complete item (ending with }, or ])
+        for end_pattern in [r'.*\}(?=\s*,?\s*$)', r'.*\](?=\s*,?\s*$)', r'.*"(?=\s*$)']:
+            pass  # fallback below
+        # Simple approach: find last complete JSON item boundary
+        # For arrays: find last "}" and close the array
+        if opener == "[":
+            last_brace = fragment.rfind("}")
+            if last_brace > 0:
+                truncated = fragment[:last_brace + 1] + "]"
+                try:
+                    return json.loads(truncated)
+                except json.JSONDecodeError:
+                    pass
+        # For objects: find last complete key-value
+        if opener == "{":
+            last_brace = fragment.rfind("}")
+            if last_brace > 0:
+                # Check if this closes a nested object — try adding outer close
+                truncated = fragment[:last_brace + 1]
+                # Count unclosed braces
+                open_count = truncated.count("{") - truncated.count("}")
+                truncated += "}" * open_count
+                try:
+                    return json.loads(truncated)
+                except json.JSONDecodeError:
+                    pass
     return None