feat(spot): 2단계 인과 기법 분석 파이프라인 마무리

common/spot/futures 경로 정비, 캔들 데이터 모듈 복원, MTF 규칙 자동 저장 및 2단계 설계·최종 정리 문서를 반영해 3단계 착수 기반을 확정한다.

Co-authored-by: Cursor <cursoragent@cursor.com>

scripts/2_run_mtf_analysis.py

@@ -20,6 +20,7 @@ from deepcoin.evaluation.mtf_report import (
     save_mtf_report,
 )
 from deepcoin.mtf.extractor import MtfFeatureExtractor
+from deepcoin.mtf.rules import derive_rules_from_report, save_mtf_rules
 from deepcoin.mtf.store import MultiTimeframeStore
 from deepcoin.techniques.runner import load_ground_truth
 
@@ -106,6 +107,9 @@ def main() -> int:
     json_path = save_mtf_report(report, settings.mtf_report_json)
     html_path = render_mtf_html(report, settings.mtf_report_html)
 
+    rule_set = derive_rules_from_report(report)
+    rules_path = save_mtf_rules(rule_set, settings.mtf_rules_json)
+
     gt = report.get("gt", {})
     top = (report.get("global_feature_ranking") or [])[:5]
     print("\n=== GT v3 MTF 상관 분석 ===")
@@ -123,6 +127,7 @@ def main() -> int:
         )
     print(f"\nJSON: {json_path}")
     print(f"HTML: {html_path}")
+    print(f"MTF rules: {rules_path}")
     return 0