Set Task Shadow Names as the Primary Task Name

[andreangelucci]

Celery tasks have the shadow_name property used to define a custom name for tasks.
When a task has multiple executions, it becomes challenging to find their task results as they share the same task names, and the task IDs are usually UUIDs. Shadow names are useful to filter and identify results for monitoring.

This PR sets the shadow name as the primary choice for the TaskResult.task_name property. If it's not set, the default task_name will be used.

Issue #468 and #379

[auvipy]

thanks for opening the PR. I will need some thorough review and thoughts on this before proceed

[auvipy]

as you can see, all the CI builds are failing due to this change. so we need proper tests cases to validate and avoid regression

[andreangelucci]

Hey, @auvipy , thanks for the feedback!
I fixed the failing unit tests by adding the missing shadow property to the mocks.

[auvipy]

This PR sets the shadow name as the primary choice for the TaskResult.task_name property. If it's not set, the default task_name will be used.

I am thinking about changing the behavior, if it has any negetive effect on existing projects?

[andreangelucci]

I am thinking about changing the behavior, if it has any negetive effect on existing projects?

@auvipy I agree, this will change the current behavior. The idea came up when I noticed that shadow names are the best way to add custom identification to a task, yet the task result lacks this information. I've added more details here: #468

Alternatively, we could create a new field in the model. What do you think about this?

[auvipy]

so our concerns here are, if this break existing behavior? and can we keep the old behavior as primary and the shadow as secondary? need more thoughts to reach on a consensus. I'm open to more discussions