feat: cross-detector language policy + tidy structure & file naming

feat: add flowchart diagrams (proses evaluasi, use case, PII) + simplify field-fit ML layer

refactor: rename detektor Responsible → Konten Berisiko (risky_content

feat: cross-field ML field-fit + context-aware typo reranking; drop Completeness

feat: add Field-Fit & Completeness detectors, rework ML layers, expand Indonesian-only lexicons

refactor: improve detectors logic, clean fallback lexicons, and polish UI layout/labels

fix: reduce evaluator false positives in prompt analysis

feat: strengthen detectors, slim API schema, and improve UX