Simplify retrieve (#420)

* parallelize validate

* use fixed annotate query param from delta to reduce retrieval to max 2 requests instead of 4

* rm redundant

* drop async request for one retrieve

* force close tcp connection

* client session per task

* retrieve schema when loading shapes, todo figure out when to call retrieve for ontologies

* retrieve schema, if fail retrieve resource

* fix flag

* rm retrieve schema specialized call after extension of delta's retrieve resource endpoint

* batch resources per client session

* cleanup

* session timeout

* rm threadpool exec

* cleanup

* lint

* bump request v

* rm retrieve schema implementation

kgforge/specializations/models/rdf_model.py

@@ -64,7 +64,7 @@
 
 DEFAULT_TYPE_ORDER = [str, float, int, bool, datetime.date, datetime.time]
 
-VALIDATION_PARALLELISM = None
+VALIDATION_PARALLELISM = 10
 
 
 class RdfModel(Model):