[Fix] Fix Inference Example, Tests, and Requirements (#5688)

* clean requirements

* modify example inference struct

* add test ci scripts

* mark test_infer as submodule

* rm deprecated cls & deps

* import of HAS_FLASH_ATTN

* prune inference tests to be run

* prune triton kernel tests

* increment pytest timeout mins

* revert import path in openmoe

tests/test_infer/test_cuda_graph.py

@@ -86,7 +86,7 @@ def run_dist(rank, world_size, port):
     check_output_consistency(128)
 
 
-@pytest.mark.dist
+@pytest.mark.largedist
 @rerun_if_address_is_in_use()
 def test_cuda_graph_infer():
     spawn(run_dist, 1)