feat: Implement new sales data analysis agents and utilities

- Added Router_de_agentes.py to manage agent interactions for sales data analysis.
- Created Analizador_de_datos_de_ventas.yaml for generating structured text reports from sales data.
- Developed Generador_sql_ventas.yaml for generating SQL queries to analyze sales data.
- Established Router_de_agente.yaml as a routing mechanism for agent requests.
- Compiled centros_disponibles.md listing available sales centers.
- Introduced primera_ejecucion_de_un_agente.py as an example for executing agents.
- Added ver_los_prompts_de_un_agente.py to inspect prompts sent to OpenAI.
- Included service account key for BigQuery access in rag-datasets-reader-sa-key.json.
- Defined schema for sales data in Objeto_ventas.json.
- Implemented utility functions for querying BigQuery in conseguir_datos_bq.py.
- Created data transformation utilities in transformar_datos.py for handling decimal and date formats.

utils/transformar_datos.py

@@ -0,0 +1,32 @@
+from decimal import Decimal    
+from datetime import datetime, date, time
+
+def convertir_decimales(obj):
+    if isinstance(obj, list):
+        return [convertir_decimales(x) for x in obj]
+    elif isinstance(obj, dict):
+        return {k: convertir_decimales(v) for k, v in obj.items()}
+    elif isinstance(obj, Decimal):
+        return float(obj)
+    else:
+        return obj
+    
+
+def convertir_fechas(obj):
+    """
+    Convierte objetos datetime, date y time a su representación ISO 8601.
+    """
+    if isinstance(obj, list):
+        return [convertir_fechas(x) for x in obj]
+    elif isinstance(obj, dict):
+        return {k: convertir_fechas(v) for k, v in obj.items()}
+    elif isinstance(obj, (datetime, date, time)):
+        return obj.isoformat()
+    else:
+        return obj
+    
+
+def limpiar_datos_para_json(data):
+    data = convertir_decimales(data)
+    data = convertir_fechas(data)
+    return data