What is the nature and description of the request?

In an environment where many jobs have been executed, deleting a project can incur in a timeout failure leading to worker thread being killed.
This problem can be avoided by increasing timeout values or by deleting old job executions via the Controller's management jobs in the administrative interface, this operation however will target all the jobs and will not be tied to a specific relationship with a project making it more of a 'last resort' type of approach.

Deleting a project could schedule a background operation that will take care of deleting its related job history entries keeping intact others, in this way being similar to the interactions that occur with inventories.

Why does the customer need this? (List the business requirements here)

It would be better not to lose all the job history to delete a single project that is no longer needed.

How would you like to achieve this? (List the functional requirements here)

Deleting a project could run a query and selectively delete only past job entries that are tied to the project itself.
If possible by re-using the logic behind the management job operations as an asynchronous task targeted to the project id.
Once the related job data has been taken care of, the actual project deletion takes place as a callback and the operation is considered completed.
Since the operation could take more than a couple of seconds or be resource intensive, while the project is being queue for deletion, an internal flag could also show the information via the API endpoints and the web interface so to not confuse the user and preventing the usage of the project.