Refactor job model and optimize job streaming.

Updated the `JobModel` with foreign key relationships and string-based status to enhance database consistency, and improved job event streaming by using `jsonable_encoder` for better serialization. Also, streamlined dependencies by adding `urllib3` to handle HTTP requests.

backend/app/models.py

@@ -7,7 +7,6 @@ from sqlalchemy import (
     JSON,
     DateTime,
     Boolean,
-    Enum,
     func,
 )
 from sqlalchemy.orm import relationship
@@ -318,8 +317,8 @@ class Jobs(Base):
     __tablename__ = "jobs"
 
     id = Column(Integer, primary_key=True, index=True)
-    experiment_parameters_id = Column(Integer, nullable=False)
-    status = Column(Enum(JobStatus), default=JobStatus.TODO, nullable=False)
-    parameters = Column(JSON, nullable=False)
+    experiment_parameters_id = Column(Integer, ForeignKey("experiment_parameters.id"))
+    status = Column(String, nullable=False)
+    parameters = relationship(ExperimentParameters)
     created_at = Column(DateTime, server_default=func.now())
     updated_at = Column(DateTime, onupdate=func.now())