fix: production readiness fixes for dependency trees and parallel agents

Critical fixes:
- Lock file TOCTOU race condition: Use atomic O_CREAT|O_EXCL for lock creation
- PID reuse vulnerability on Windows: Store PID:CREATE_TIME in lock file to
  detect when a different process has reused the same PID
- WAL mode on network drives: Detect network paths (UNC, mapped drives, NFS,
  CIFS) and fall back to DELETE journal mode to prevent corruption

High priority fixes:
- JSON migration now preserves dependencies field during legacy migration
- Process tree termination on Windows: Use psutil to kill child processes
  recursively to prevent orphaned browser instances
- Retry backoff jitter: Add random 30% jitter to prevent synchronized retries
  under high contention with 5 concurrent agents

Files changed:
- server/services/process_manager.py: Atomic lock creation, PID+create_time
- api/database.py: Network filesystem detection for WAL mode fallback
- api/migration.py: Add dependencies field to JSON migration
- parallel_orchestrator.py: _kill_process_tree helper function
- mcp_server/feature_mcp.py: Add jitter to exponential backoff

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

mcp_server/feature_mcp.py

@@ -20,6 +20,7 @@ Tools:
 
 import json
 import os
+import random
 import sys
 import threading
 import time as _time
@@ -313,9 +314,11 @@ def _feature_claim_next_internal(attempt: int = 0) -> str:
             if result.rowcount == 0:
                 # Another process claimed it first - retry with backoff
                 session.close()
-                # Exponential backoff: 0.1s, 0.2s, 0.4s, ... up to 1.0s
+                # Exponential backoff with jitter: base 0.1s, 0.2s, 0.4s, ... up to 1.0s
+                # Jitter of up to 30% prevents synchronized retries under high contention
                 backoff = min(0.1 * (2 ** attempt), 1.0)
-                _time.sleep(backoff)
+                jitter = random.uniform(0, backoff * 0.3)
+                _time.sleep(backoff + jitter)
                 return _feature_claim_next_internal(attempt + 1)
 
             # Fetch the claimed feature