[ CI ] Fix Failing Magic Wand Test (#311)

SUMMARY:
* Fix https://app.asana.com/0/1206976017967941/1207600124500657
* Issue was that we were not tearing down the model fully before
initializing a new vllm instance. Calling `gc.collect()` seems to force
this to happen

Author: robertgshaw2-redhat

tests/models_core/test_magic_wand.py

@@ -3,9 +3,11 @@
 As a result, in this test, we just confirm that the top selected tokens of the 
 sparse models are in the top N selections of same model running dense.
 
-Run `pytest tests/models/test_compressed.py`.
+Run `pytest tests/models_core/test_magic_wand.py`.
 """
 
+import gc
+
 import pytest
 
 from tests.models.utils import check_logprobs_close
@@ -47,6 +49,7 @@ def test_magic_wand(
     dense_outputs = dense_model.generate_greedy_logprobs(
         example_prompts, max_tokens, num_logprobs)
     del dense_model
+    gc.collect()
 
     sparse_model = vllm_runner(
         model_name=model_name,