[DOP-38389] Spark job.name=unknown should be detached from others

[dolfinus]

Change Summary

Due to OpenLineage/OpenLineage#3846 RunEvents from Openlineage integration for Spark may contain job.name="unknown" (eventType=STARTED), but latter RunEvents (RUNNING, SUCCESS and so on) are received with proper job name (e.g. job.name="my_session").

This creates temporary job in DataRentgen, which is constantly updated:

UPDATE job SET parent_job_id=234 WHERE job_id=123; -- from one event
UPDATE job SET parent_job_id=345 WHERE job_id=123; -- from another event
UPDATE job SET parent_job_id=456 WHERE job_id=123; -- from third event

Same for job tags and dependencies.

Now consumer for case job.name="unknown" resets parent job, dependencies and tag values, so UPDATE queries are not emitted. Added DB migration to reset those in databaset as well.

Related issue number

Checklist

• Commit message and PR title is comprehensive
• Keep the change as small as possible
• Unit and integration tests for the changes exist
• Tests pass on CI and coverage does not decrease
• Documentation reflects the changes where applicable
• docs/changelog/next_release/<pull request or issue id>.<change type>.rst file added describing change
  (see CONTRIBUTING.rst for details.)
• My PR is ready to review.

[github-actions]

Coverage Report •

File	Stmts	Miss	Branch	BrPart	Cover	Missing
data_rentgen/consumer/extractors/impl
spark.py	68	1	16	2	96%	93->95, 113->114, 114
TOTAL	7710	518	1306	211	92%