[MTP] follow custom deepseek modeling changes to support graph mode (#636)

<!--  Thanks for sending a pull request!

BEFORE SUBMITTING, PLEASE READ
https://docs.vllm.ai/en/latest/contributing/overview.html

-->
### What this PR does / why we need it?

As custom deepseek modeling do some changes to support graph mode in
https://github.com/vllm-project/vllm-ascend/pull/585, so i follow it to
change custom deepseek_mtp modeling.

And some modifications for k>1 were not carried over by the
https://github.com/vllm-project/vllm-ascend/pull/429, now i add it.

In order to better take care of the MTP feature in the vllm-ascend
repository, I added cases related to graph mode(torchair), but i skip it
since torchair can not correctly clean up memory in vllmrunner.

Also i add some case for MTP quantization weights, but test weight is
not ready, so i skip it and i will open it when test quant weights is
ready.

https://github.com/vllm-project/vllm-ascend/pull/648 did not completely
fix the sample
change(https://github.com/vllm-project/vllm-ascend/issues/660) issue, I
added the relevant changes.

### Does this PR introduce _any_ user-facing change?
now, u can use following method to use mtp in deepseek v3/r1 float or
quant weights with eager mode.
```python
llm = LLM(
    model="wemaster/deepseek_mtp_main_random_bf16",
    tensor_parallel_size=2,
    speculative_config={
        "num_speculative_tokens": 1,
    },
    enforce_eager=True,
    trust_remote_code=True,
    disable_log_stats=False,
    gpu_memory_utilization=0.8,
    max_model_len=64,
)
```

or use mtp in deepseek v3/r1 float or quant weights with graph
mode（torchair）
```python
llm = LLM(
    model="wemaster/deepseek_mtp_main_random_bf16",
    tensor_parallel_size=2,
    speculative_config={
        "num_speculative_tokens": 1,
    },
    trust_remote_code=True,
    additional_config={
        'enable_graph_mode': True,
    },
    disable_log_stats=False,
    gpu_memory_utilization=0.8,
    max_model_len=64,
)
```

add notes:
1. now, we support k>1, so u can set num_speculative_tokens > 1 if there
is sufficient redundant computing power;
2. MTP is not supported in V1, we will support it when vLLM does it in
https://github.com/vllm-project/vllm/issues/13500.
3. if u run MTP failed by `segmentation fault`, u can follow v0.7.3
patch https://github.com/vllm-project/vllm-ascend/pull/236 file
`vllm_ascend/patch/patch_metrics.py` method
`__npu_async_metrics_collector_init__`

### How was this patch tested?
local tested passed and test by CI

Signed-off-by: mengwei805 <mengwei25@huawei.com>

tests/singlecard/spec_decode/e2e/conftest.py

@@ -17,7 +17,9 @@
 # limitations under the License.
 #
 
+import shutil
 from itertools import cycle
+from pathlib import Path
 from typing import List, Optional, Sequence, Tuple, Union
 
 import pytest
@@ -177,6 +179,12 @@ def _check_logprobs_when_output_disabled(
         assert spec_pos_logprob_token_id in baseline_pos_logprobs
 
 
+def _clean_torchair_cache():
+    cache_path = Path.cwd() / '.torchair_cache'
+    if cache_path.exists() and cache_path.is_dir():
+        shutil.rmtree(cache_path)
+
+
 def run_equality_correctness_test(
         vllm_runner,
         common_llm_kwargs,
@@ -219,10 +227,20 @@ def run_equality_correctness_test(
                                      logprobs=logprobs,
                                      prompt_logprobs=prompt_logprobs)
 
+    # TODO current torchair graph mode needs clean torchair cache.
+    # if do not clean, it will raise error
+    additional_config = common_llm_kwargs.get("additional_config")
+    enable_graph_mode = additional_config.get(
+        "enable_graph_mode") if additional_config else False
+
     with vllm_runner(**org_args) as vllm_model:
+        if enable_graph_mode:
+            _clean_torchair_cache()
         org_outputs = vllm_model.generate_w_logprobs(prompts, sampling_params)
 
     with vllm_runner(**sd_args) as vllm_model:
+        if enable_graph_mode:
+            _clean_torchair_cache()
         if ensure_all_accepted or expected_acceptance_rate is not None:
             # Force log interval to be 0 to catch all metrics.
             stat_logger = vllm_model.model.llm_engine.stat_loggers[