[CI] Refactor CI (#952)

1. remove some useless test func and file
2. fix format.sh problem
3. enable full test for singlecard and multicard
4. move long term test to long_term folder. For this kind of test, it
only runs by labeled and daily test. Include: spec decode、accuracy test

## After refactor:
There are 4 test modules
- `singlecard`: contains the test running on one NPU. It'll be run for
each PR and daily test.
- `multicard`: contains the test running on multi NPUs. It'll be run for
each PR and daily test.
- `long_term`: contains the test that cost much time(Now include `spec
decode` and `accuracy` test). It'll be run for the PR with
`long-term-test` labeled and daily test.
- `e2e`: contains the test for doc and pd feature. It'll be run for the
PR with `pd-test` labeled and daily test.

## Todo:
1. some test are skipped, they should be fixed and reenabled in the
future.
2. pyhccl test for multicard doesn't work at all. It should be enabled
as well.
3. ensure long-term-test pass by daily test.

### Know issue
Now, `ready` labels is required to start pd test or long term test. And
when `long-term-test` or `pd-test` is labeled after another one, the old
labeled test will be re-run again. So the labeled test should be ran in
the following step:

1. decide which test need run, then label it. `long-term-test` or
`pd-test` or both.
2. add `ready-for-test` label, then the test will be ran.

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

tests/model_utils.py

@@ -20,9 +20,6 @@
 import warnings
 from typing import Callable, Dict, List, Optional, Sequence, Tuple, Union
 
-import torch
-from vllm.config import ModelConfig, TaskOption
-from vllm.inputs import InputContext
 from vllm.sequence import Logprob, PromptLogprobs, SampleLogprobs
 
 TokensText = Tuple[List[int], str]
@@ -264,45 +261,6 @@ def check_logprobs_close(
                     warnings.warn(fail_msg, stacklevel=2)
 
 
-def build_model_context(model_name: str,
-                        task: TaskOption = "auto",
-                        tokenizer_name: Optional[str] = None,
-                        trust_remote_code: bool = False,
-                        dtype: Optional[Union[str, torch.dtype]] = None,
-                        mm_processor_kwargs: Optional[Dict] = None,
-                        limit_mm_per_prompt: Optional[Dict] = None):
-    """Creates an InputContext for a given model.
-
-    Args:
-        model_name: Name of the model being considered.
-        tokenizer_name: Name of the tokenizer being considered.
-        trust_remote_code: Whether or not to allow loading remote code.
-        mm_processor_kwargs: optional processor kwargs for to be leveraged
-            in the input processor, mapper, dummy data creation, etc.
-        limit_mm_per_prompt: Multimodal limits.
-
-    Returns:
-        InputContext for the model being considered.
-    """
-    if tokenizer_name is None:
-        tokenizer_name = model_name
-    if dtype is None:
-        dtype = "half"
-
-    model_config = ModelConfig(
-        model_name,
-        task=task,
-        tokenizer=tokenizer_name,
-        tokenizer_mode="auto",
-        trust_remote_code=trust_remote_code,
-        dtype=dtype,
-        seed=0,
-        mm_processor_kwargs=mm_processor_kwargs,
-        limit_mm_per_prompt=limit_mm_per_prompt,
-    )
-    return InputContext(model_config)
-
-
 def qwen_prompt(questions: List[str]) -> List[str]:
     placeholder = "<|image_pad|>"
     return [("<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n"
@@ -313,4 +271,4 @@ def qwen_prompt(questions: List[str]) -> List[str]:
 # Map of prompt templates for different models.
 PROMPT_TEMPLATES: dict[str, Callable] = {
     "qwen2.5vl": qwen_prompt,
-}
+}