[maintenance] Centralize pagination input validation tests

[PGijsbers]

The list tasks PR introduced some tests to assert the input validation of the Pagination dependency.
There may be other tests which test this behavior, I haven't checked.
There should be one central set of tests which test this input validation.

Note, the central tests are limited to input validation, they cannot and should not test whether the pagination parameters are used correctly. That's endpoint specific.

@saathviksheerla will take on this issue, unless they indicate otherwise.

[saathviksheerla]

I am willing to work on this issue and will raise a PR soon. Thanks for assigning the issue to me !

[saathviksheerla]

Hi @PGijsbers , I have 2 quick clarifying questions before I proceed with implementation. Please help me get a clear understanding.

• Self-vallidating Model:
  instead of just centralizing the tests, would you like me to move the validation logic into pagination pydantic model in dependencies.py using Field(ge=0). But we have to note that using Field means the API will return Unprocessable Entity for negative values, whereas at some parts of code(my previous implemetation) silently fixes it using max(0, offset). If you agree with this approach, should I also set a global max limit (e.g. le=1000) at the model level?

• Refactoring existing endpoints:
  If we update the model, should I also remove the manual max(0, ...) checks from /tasks/list and /datasets/list so we have a single source of truth for pagination rules?

I’m already clear on creating the dedicated standalone tests for these validation rules that will run independently of any specific endpoint
Thanks!