test: trim agentic eval bucket tests to essential coverage

Collapse fixtures into module constants, merge the accept/reject checks
into one test, and drop redundant cases; added test lines go from 212 to
100 with the two core guards kept (agentic rows validate only in
agentic_evals; realistic rows split end-to-end through the real schema).

中文：精简 agentic 评估分桶测试——fixture 合并为模块常量，接受/拒绝检查合并
为单个测试并移除冗余用例，新增测试行数从 212 降至 100，保留两项核心守卫
（agentic 行仅能通过 agentic_evals 校验；真实行形状端到端经过真实 schema
拆分）。

Author: Oseltamivir

utils/matrix_logic/test_validation.py

@@ -1405,108 +1405,44 @@ def test_scenario_type_must_be_nonempty_and_supported(self, scenario_type):
 # Test ChangelogMatrixEntry
 # =============================================================================
 
-@pytest.fixture
-def valid_agentic_eval_row():
-    """Agentic SWE-bench eval row as emitted by generate_sweep_configs --evals-only."""
-    return {
-        "image": "vllm/vllm-openai:nightly-dev-x86_64-cu13.0.1-904e4ec",
-        "model": "deepseek-ai/DeepSeek-V4-Pro",
-        "model-prefix": "dsv4",
-        "precision": "fp4",
-        "framework": "vllm",
-        "runner": "cluster:b300-nv",
-        "tp": 8,
-        "pp": 1,
-        "dcp-size": 1,
-        "pcp-size": 1,
-        "ep": 8,
-        "dp-attn": True,
-        "spec-decoding": "mtp",
-        "conc": 224,
-        "kv-offloading": "none",
-        "router": {"name": "vllm-router", "version": "0.1.14"},
-        "total-cpu-dram-gb": 0,
-        "duration": 3600,
-        "exp-name": "dsv4_tp8_conc224_kvnone_spec-mtp",
-        "scenario-type": "agentic-coding",
-        "run-eval": True,
-        "eval-only": True,
-    }
-
-
-@pytest.fixture
-def valid_changelog_metadata():
-    return {
-        "base_ref": "base",
-        "head_ref": "head",
-        "entries": [{
-            "config-keys": ["test-config"],
-            "description": ["Test entry"],
-            "pr-link": "https://github.com/SemiAnalysisAI/InferenceX/pull/1",
-        }],
-    }
+AGENTIC_EVAL_ROW = {
+    "image": "vllm/vllm-openai:nightly", "model": "deepseek-ai/DeepSeek-V4-Pro",
+    "model-prefix": "dsv4", "precision": "fp4", "framework": "vllm",
+    "runner": "cluster:b300-nv", "tp": 8, "pp": 1, "dcp-size": 1,
+    "pcp-size": 1, "ep": 8, "dp-attn": True, "spec-decoding": "mtp",
+    "conc": 224, "kv-offloading": "none", "total-cpu-dram-gb": 0,
+    "duration": 3600, "exp-name": "dsv4_tp8_conc224_kvnone_spec-mtp",
+    "scenario-type": "agentic-coding", "run-eval": True, "eval-only": True,
+}
+
+CHANGELOG_METADATA = {
+    "base_ref": "base", "head_ref": "head",
+    "entries": [{
+        "config-keys": ["test-config"], "description": ["Test entry"],
+        "pr-link": "https://github.com/SemiAnalysisAI/InferenceX/pull/1",
+    }],
+}
 
 
 class TestChangelogMatrixEntry:
     """Tests for the final search-space contract consumed by run-sweep.yml."""
 
-    def test_agentic_eval_rows_validate_in_agentic_evals(
-        self, valid_agentic_eval_row, valid_changelog_metadata,
-    ):
+    def test_agentic_eval_rows_live_in_agentic_evals_only(self):
+        """`evals` is dispatched with fixed-seq-len inputs (isl/osl/
+        max-model-len), so agentic rows must only validate in agentic_evals."""
         entry = ChangelogMatrixEntry.model_validate({
-            "agentic_evals": [valid_agentic_eval_row],
-            "changelog_metadata": valid_changelog_metadata,
+            "agentic_evals": [AGENTIC_EVAL_ROW],
+            "changelog_metadata": CHANGELOG_METADATA,
         })
-
-        assert entry.agentic_evals[0].scenario_type == "agentic-coding"
         assert entry.agentic_evals[0].run_eval is True
         assert entry.agentic_evals[0].eval_only is True
 
-    def test_evals_bucket_rejects_agentic_rows(
-        self, valid_agentic_eval_row, valid_changelog_metadata,
-    ):
-        """The `evals` bucket is dispatched with fixed-seq-len inputs
-        (isl/osl/max-model-len), so agentic rows must never validate there."""
         with pytest.raises(ValueError):
             ChangelogMatrixEntry.model_validate({
-                "evals": [valid_agentic_eval_row],
-                "changelog_metadata": valid_changelog_metadata,
+                "evals": [AGENTIC_EVAL_ROW],
+                "changelog_metadata": CHANGELOG_METADATA,
             })
 
-    def test_evals_bucket_accepts_fixed_seq_len_rows(
-        self, valid_single_node_matrix_entry, valid_changelog_metadata,
-    ):
-        eval_row = {
-            **valid_single_node_matrix_entry,
-            "run-eval": True,
-            "eval-only": True,
-        }
-        entry = ChangelogMatrixEntry.model_validate({
-            "evals": [eval_row],
-            "changelog_metadata": valid_changelog_metadata,
-        })
-
-        assert entry.evals[0].run_eval is True
-        assert entry.evals[0].eval_only is True
-
-    def test_agentic_benchmark_rows_dump_without_eval_flags(
-        self, valid_agentic_eval_row, valid_changelog_metadata,
-    ):
-        """Benchmark rows omit run-eval/eval-only, and exclude_none must keep
-        them out of the dump so the dispatched row shape is unchanged."""
-        benchmark_row = {
-            k: v for k, v in valid_agentic_eval_row.items()
-            if k not in ("run-eval", "eval-only")
-        }
-        entry = ChangelogMatrixEntry.model_validate({
-            "single_node": {"agentic": [benchmark_row]},
-            "changelog_metadata": valid_changelog_metadata,
-        })
-        dumped = entry.model_dump(by_alias=True, exclude_none=True)
-
-        assert "run-eval" not in dumped["single_node"]["agentic"][0]
-        assert "eval-only" not in dumped["single_node"]["agentic"][0]
-
 
 # =============================================================================
 # Test load_config_files

utils/test_process_changelog.py

@@ -599,91 +599,43 @@ def test_eval_rows_split_into_fixed_and_agentic_buckets(
     - Mixed fixed-seq-len and agentic eval selection
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1
 """
-
-    fixed_eval_row = {
+    common = {
         "image": "vllm/vllm-openai:v0.11.0",
-        "model": "deepseek-ai/DeepSeek-V4-Pro",
-        "model-prefix": "dsv4",
-        "precision": "fp4",
-        "framework": "vllm",
-        "spec-decoding": "mtp",
-        "runner": "cluster:b300-nv",
-        "isl": 8192,
-        "osl": 1024,
-        "tp": 8,
-        "pp": 1,
-        "dcp-size": 1,
-        "pcp-size": 1,
-        "ep": 8,
-        "dp-attn": True,
-        "conc": 224,
-        "max-model-len": 10240,
-        "exp-name": "dsv4_8k1k_fixed_eval",
-        "disagg": False,
-        "run-eval": True,
-        "eval-only": True,
+        "model": "deepseek-ai/DeepSeek-V4-Pro", "model-prefix": "dsv4",
+        "precision": "fp4", "framework": "vllm", "spec-decoding": "mtp",
+        "runner": "cluster:b300-nv", "tp": 8, "pp": 1, "dcp-size": 1,
+        "pcp-size": 1, "ep": 8, "dp-attn": True, "conc": 224,
+        "run-eval": True, "eval-only": True,
+    }
+    fixed_eval_row = {
+        **common, "isl": 8192, "osl": 1024, "max-model-len": 10240,
+        "disagg": False, "exp-name": "fixed_eval",
     }
     agentic_eval_row = {
-        "image": "vllm/vllm-openai:v0.11.0",
-        "model": "deepseek-ai/DeepSeek-V4-Pro",
-        "model-prefix": "dsv4",
-        "precision": "fp4",
-        "framework": "vllm",
-        "runner": "cluster:b300-nv",
-        "tp": 8,
-        "pp": 1,
-        "dcp-size": 1,
-        "pcp-size": 1,
-        "ep": 8,
-        "dp-attn": True,
-        "spec-decoding": "mtp",
-        "conc": 224,
-        "kv-offloading": "none",
-        "router": {"name": "vllm-router", "version": "0.1.14"},
-        "total-cpu-dram-gb": 0,
-        "duration": 3600,
-        "exp-name": "dsv4_agentic_eval",
-        "scenario-type": "agentic-coding",
-        "run-eval": True,
-        "eval-only": True,
+        **common, "kv-offloading": "none", "total-cpu-dram-gb": 0,
+        "duration": 3600, "scenario-type": "agentic-coding",
+        "exp-name": "agentic_eval",
     }
 
     monkeypatch.setattr(
-        process_changelog,
-        "get_added_lines",
-        lambda *_: added_yaml,
-    )
+        process_changelog, "get_added_lines", lambda *_: added_yaml)
     monkeypatch.setattr(
-        process_changelog,
-        "load_config_files",
-        lambda _: {"test-config": {}},
-    )
+        process_changelog, "load_config_files", lambda _: {"test-config": {}})
 
     def fake_run(command, **kwargs):
-        if "--evals-only" in command:
-            stdout = json.dumps([fixed_eval_row, agentic_eval_row])
-        else:
-            stdout = "[]"
-        return SimpleNamespace(stdout=stdout)
+        is_evals = "--evals-only" in command
+        rows = [fixed_eval_row, agentic_eval_row] if is_evals else []
+        return SimpleNamespace(stdout=json.dumps(rows))
 
     monkeypatch.setattr(subprocess, "run", fake_run)
     monkeypatch.setattr(sys, "argv", [
-        "process_changelog.py",
-        "--base-ref", "base",
-        "--head-ref", "head",
+        "process_changelog.py", "--base-ref", "base", "--head-ref", "head",
         "--changelog-file", "perf-changelog.yaml",
     ])
 
     process_changelog.main()
 
     output = json.loads(capsys.readouterr().out)
-    assert [row["exp-name"] for row in output["evals"]] == [
-        "dsv4_8k1k_fixed_eval"
-    ]
-    assert [row["exp-name"] for row in output["agentic_evals"]] == [
-        "dsv4_agentic_eval"
-    ]
+    assert [r["exp-name"] for r in output["evals"]] == ["fixed_eval"]
+    assert [r["exp-name"] for r in output["agentic_evals"]] == ["agentic_eval"]
     assert output["multinode_evals"] == []
-    assert output["agentic_evals"][0]["scenario-type"] == "agentic-coding"
-    assert output["agentic_evals"][0]["run-eval"] is True
-    assert output["agentic_evals"][0]["eval-only"] is True