feat: implement upsert logic and automatic student resolution for financial records

feat: restrict bulk data ingestion by role and implement payload filtering for data security

refactor: remove manual migration script and handle NaN values in data processing

feat: integrate automatic database migration execution on application startup

feat: implement upsert logic for fact tables, update CORS settings, and extend estudiante models with enrollment fields

feat: add dynamic DimTiempo resolution and support for projected collections in financial processing

feat: implement dynamic budget category creation based on program names during data ingestion

feat: initialize hechos_finance list in data processing pipeline

feat: add financial reporting fields to schema, update data ingestion logic, and lower AI confidence threshold for revision

feat: expand source type scope, optimize teacher resolution logic, and improve MLOps feedback handling with pending log prioritization.

feat: implement marketing data ingestion with normalization and add bulk deduplication/cleaning to financial records

feat: add DimCategoriaFinanciera model import to app.py

feat: implement automatic creation of default financial category for budget payloads in app sync process

refactor: optimize bulk data processing by caching dimensions and batching database operations

feat: expand data ingestion models to support marketing and survey pipelines and centralize fact insertion logic

feat: expandir endpoints de dashboard con 6 KPIs ejecutivos y analíticas de riesgo académico detalladas

refactor: update student fuzzy matching logic to use find_best_match with entity objects

feat: implement student resolution logic with fuzzy matching and automated record creation in financial ingestions

fix: add error handling and graceful fallbacks for dashboard database queries to prevent failures on missing tables

feat: import additional fact models for academic and financial analysis in app.py

feat: add CMI dashboard endpoints and integrate fuzzy matching suggestions for student records

Fix NoneType error: Always create DimEstudiante regardless of NLP confidence

Add demographic fields extraction and dynamic dimension updates

Fix FK constraint on NLP audit log

Fix NameError in dimension caching

Fix database statement timeout during batch processing

feat: implement batch NLP analysis endpoint with automated entity resolution and data integration logic.

refactor: loosen payload constraints and introduce conditional validation for non-financial data sources

feat: import Field and field_validator from pydantic in app.py

feat: implement financial unpivot ingestion endpoint and add new dimension and fact tables to schema

feat: integrate MLOps memory lookup and fuzzy matching into nlp analysis endpoint

chore: update password hashing algorithm to pbkdf2_sha256 and relax CORS middleware configurations

refactor: rename analista_datos role to analista_datos_marketing and remove automatic table creation

feat: implement JWT-based authentication and role-based access control, and add LogAuditoriaNlp table for MLOps feedback tracking

refactor: remove auditorias count from academic performance statistics query

feat: implement Jaro-Winkler fuzzy matching for student linking and add data quality diagnostic endpoint

feat: implement security hardening, CORS middleware, bulk ingestion, and protected diagnostic endpoint

Diag: Endpoint /api/v1/diagnostico para trazabilidad de tablas en Supabase

Fix: Alinear modelos SQLAlchemy con DDL real de Supabase

Fix: Alinear modelos SQLAlchemy con DDL real de Supabase

Feat: Endpoint OLAP para cruce de riesgos académicos

Feat: Endpoint OLAP para cruce de riesgos académicos

Fix: Transacciones seguras con db.flush()

Fix: Autogeneración de Dimensiones Maestras para evitar Error 500 por Foreign Keys

Feat: Endpoint GET para consulta cruzada de riesgos académicos

Fix: Eliminar importación no utilizada de engine para estabilizar ASGI