[CI]: Add core_model and cpu markers for L1 use case. (#1709)

Signed-off-by: zhumingjue <zhumingjue@huawei.com>

docs/contributing/ci/CI_5levels.md

@@ -67,8 +67,8 @@ Through five levels (L1-L5) and common (Common) specifications, the system clari
       <td>Online & Offline (multiple deployment scenarios):<br>real model, normal inference function, normal accuracy</td>
       <td>&lt;30min</td>
       <td>
-        /tests/e2e/online_serving/test_{model_name}_expansion.py<br>
-        /tests/e2e/offline_inference/test_{model_name}_expansion.py
+        /tests/e2e/online_serving/test_{model_name}.py<br>
+        /tests/e2e/offline_inference/test_{model_name}.py
       </td>
       <td>
         <a href="#chapter-2-l3-level-testing-core-integration-performance-and-accuracy-verification">Chapter 2</a><br>

tests/diffusion/distributed/test_hsdp.py

@@ -11,7 +11,7 @@ import torch.nn as nn
 from vllm_omni.diffusion.data import DiffusionParallelConfig
 from vllm_omni.diffusion.distributed.hsdp import HSDPInferenceConfig
 
-pytestmark = [pytest.mark.diffusion, pytest.mark.parallel, pytest.mark.cpu]
+pytestmark = [pytest.mark.diffusion, pytest.mark.parallel, pytest.mark.cpu, pytest.mark.core_model]
 
 
 class TestHSDPInferenceConfig:

tests/diffusion/distributed/test_parallel_state_sp_groups.py

@@ -11,10 +11,7 @@ import torch
 
 from vllm_omni.diffusion.distributed.parallel_state import RankGenerator, set_seq_parallel_pg
 
-pytestmark = [
-    pytest.mark.diffusion,
-    pytest.mark.parallel,
-]
+pytestmark = [pytest.mark.diffusion, pytest.mark.parallel, pytest.mark.core_model, pytest.mark.cpu]
 
 
 def _fake_new_group_factory(created_groups: list[SimpleNamespace]):

tests/diffusion/distributed/test_sp_plan_hooks.py

@@ -47,6 +47,8 @@ requires_distributed = pytest.mark.skipif(
 pytestmark = [
     pytest.mark.diffusion,
     pytest.mark.parallel,
+    pytest.mark.core_model,
+    pytest.mark.cpu,
 ]
 
 # =============================================================================

tests/diffusion/distributed/test_vae_patch_parallel.py

@@ -8,6 +8,8 @@ import torch
 
 from vllm_omni.diffusion.distributed import vae_patch_parallel as vae_patch_parallel
 
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
 
 class _DummyConfig:
     def __init__(self, **attrs):

tests/diffusion/models/nextstep_1_1/test_nextstep_cfg_parallel_layout.py

@@ -8,6 +8,8 @@ import vllm_omni.diffusion.models.nextstep_1_1.pipeline_nextstep_1_1 as nextstep
 from vllm_omni.diffusion.models.nextstep_1_1.modeling_nextstep_heads import FlowMatchingHead
 from vllm_omni.diffusion.models.nextstep_1_1.pipeline_nextstep_1_1 import NextStep11Pipeline
 
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
 
 class _DummyImageHead:
     def __init__(self, token_dim: int):

tests/diffusion/models/z_image/test_zimage_tp_constraints.py

@@ -2,6 +2,8 @@ import pytest
 
 from vllm_omni.diffusion.models.z_image.z_image_transformer import validate_zimage_tp_constraints
 
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
 
 def test_validate_zimage_tp_constraints_tp2_ok():
     ffn_hidden_dim, final_out_dims, supported_tp = validate_zimage_tp_constraints(

tests/diffusion/test_multiproc_executor_concurrency.py

@@ -12,7 +12,7 @@ from vllm_omni.diffusion.data import DiffusionOutput
 from vllm_omni.diffusion.executor.multiproc_executor import MultiprocDiffusionExecutor
 from vllm_omni.diffusion.scheduler import Scheduler
 
-pytestmark = [pytest.mark.diffusion]
+pytestmark = [pytest.mark.diffusion, pytest.mark.core_model, pytest.mark.cpu]
 
 
 # ───────────────────────────────────────────── helpers ─────────────────────

tests/distributed/omni_connectors/test_mooncake_transfer_engine_buffer.py

@@ -17,7 +17,7 @@ from vllm_omni.distributed.omni_connectors.connectors.mooncake_transfer_engine_c
 )
 
 # All tests in this file are pure-CPU unit tests for the memory allocator.
-pytestmark = [pytest.mark.cpu, pytest.mark.parallel]
+pytestmark = [pytest.mark.cpu, pytest.mark.parallel, pytest.mark.core_model]
 
 
 @pytest.mark.core_model

tests/distributed/omni_connectors/test_omni_connector_configs.py

@@ -8,6 +8,8 @@ import pytest
 # Use the new import path for initialization utilities
 from vllm_omni.distributed.omni_connectors.utils.initialization import load_omni_transfer_config
 
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
 
 def get_config_files():
     """Helper to find config files."""

tests/model_executor/models/qwen2_5_omni/test_qwen2_5_omni_embed.py

@@ -19,6 +19,8 @@ from vllm.model_executor.models.qwen2_5_omni_thinker import (
     merge_interleaved_embeddings,
 )
 
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
 # Fake token IDs
 AUDIO_TOKEN_ID = 1001
 IMAGE_TOKEN_ID = 1002