fix(hybrid): apply_adjustment에서 역사 행(yhat=None) 산술 오류 수정

fix(hybrid): 차트에 역사 데이터(실측값) 표시 안 되는 문제 수정

feat(hybrid): LLM 보정 범위를 앙상블 불확실성 기반 동적 클램프로 교체

feat: Add 80% confidence interval support for Prophet, ARIMA, and Theta models

fix(data_service): EUC-KR 인코딩 및 한국어 날짜 형식 지원 추가

fix(backend): strip markdown code blocks from Claude response in timesfm_service

fix: install timesfm from GitHub source for 2.5 support, restore correct API

fix: use timesfm-1.0-200m-pytorch (PyPI compatible), cap horizon at 128

fix: use correct timesfm.TimesFm API instead of non-existent TimesFM_2p5_200M_torch

fix: AI interpret response check, remove 95% CI option, serialize dict to JSON for Claude prompt

fix: address code review issues in timesfm_service.py

feat: add timesfm_service.py with singleton model loading

feat(hybrid): Add data polarity option and Excel export functionality

chore: fix python 3.9 type hints, integrate venv in start.sh, and add dotenv support

fix: resolve macOS OpenMP conflict between PyTorch and LightGBM in ensemble

fix: add KeyError guard and ImportError handling in llm_adjustment_service

feat: add llm_adjustment_service with Claude API integration and tests

fix: use get_running_loop, add length guard and DataFrame type check in ensemble_service

feat: add ensemble_service with async orchestrator and pure-function tests

fix: resolve validation split causing insufficient training samples in N-BEATS

Add models:LightGBM,Theta, ETS

Creating a new common data profiling model, etc.

Add ARIMA/SARIMA models

Add Prophet, XGBoost models

first commit