[CI] disable glm4.1v and fix the flashinfer installation (#11902)

Signed-off-by: Shangming Cai <csmthu@gmail.com>

test/srt/run_suite.py

@@ -133,7 +133,7 @@ suites = {
         TestFile("test_triton_sliding_window.py", 250),
         TestFile("test_utils_update_weights.py", 48),
         TestFile("test_vision_chunked_prefill.py", 175),
-        TestFile("test_vision_openai_server_a.py", 608),
+        TestFile("test_vision_openai_server_a.py", 918),
         TestFile("test_vlm_input_format.py", 300),
     ],
     "per-commit-2-gpu": [

test/srt/test_vision_openai_server_a.py

@@ -136,6 +136,9 @@ class TestKimiVLServer(ImageOpenAITestMixin):
         pass
 
 
+@unittest.skip(
+    "Temporarily disabling this test to fix CI. It should be re-enabled when #11800 is done."
+)
 class TestGLM41VServer(ImageOpenAITestMixin, VideoOpenAITestMixin):
     model = "zai-org/GLM-4.1V-9B-Thinking"
     extra_args = [