[TESTS] Add global random seed for deterministic test

[Stephen0512]

Description

This PR sets a fixed global random seed after imports in the test file to ensure deterministic and reproducible test behavior.

Motivation

Some tests rely on randomness without a fixed seed, which can lead to flaky tests, inconsistent CI results, and difficulty reproducing failures.

Change

A global random seed is initialized to make random behavior consistent across test runs.

Benefits

• More reproducible and reliable tests
• Easier debugging of failures
• Consistent CI behavior

The issue was identified during an ongoing research project.

[Copilot]

Pull request overview

This PR aims to make tests/test_hawkular.py deterministic by fixing the randomness source so that sampling-based test setup is reproducible across runs.

Changes:

• Imports seed from Python’s random module.
• Seeds the global RNG (seed(17)) at module import time.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Calling seed(17) at module import time mutates Python’s global RNG state for the entire pytest process, which can create cross-test coupling and order-dependent behavior (this module may be imported mid-run, resetting the RNG sequence for other tests). Prefer using a local RNG (e.g., random.Random(17).sample(...)) or save/restore random.getstate() / random.setstate() around the specific sampling so the determinism is limited to this test logic. If the intent is to seed the whole suite, do it once in a session-scoped autouse fixture (e.g., in tests/conftest.py) rather than in a single test module import.

[ogajduse]

This suggestion seems reasonable. @Stephen0512, mind looking into it?

[Stephen0512]

Hello @ogajduse, thanks for taking a look at this PR. I have changed the code to make the randomness local.