Bulk task deletion must be atomic across running-task checks

[ionfwsrijan]

Problem

Bulk deletion can partially delete completed tasks even when the request should fail because one selected task is still running.

Evidence

Affected areas:

• backend/secuscan/routes.py
• testing/backend/integration/test_task_cleanup.py
• testing/backend/integration/test_bulk_delete_contract.py

The expected contract is: if any selected task is running, reject the entire request and delete nothing.

Impact

Partial deletion can erase historical reports/findings while leaving running tasks behind, corrupting dashboard state and audit history.

Expected fix

• Fetch all selected tasks first
• Validate all IDs exist and none are running before deleting anything
• Perform deletion in a transaction
• Return consistent { success, deleted_count } response shape
• Add regression tests for mixed completed/running task lists

[ionfwsrijan]

@utksh1 You may review and assign me this issue

[ariesshadow]

I’d like to work on this issue. Please assign it to me if still available. I will also add regression tests for mixed running task sets.