[Bugfix][V1] Fix deepseek with v1 (#958)

### What this PR does / why we need it?
Fix deepseek with v1, this error is introdeced by
https://github.com/vllm-project/vllm-ascend/pull/945. and this pr fix
the block table of mla

### How was this patch tested?
CI passed with new addedtest.

Signed-off-by: Mengqing Cao <cmq0113@163.com>

tests/multicard/test_offline_inference_distributed.py

@@ -22,7 +22,6 @@ Run `pytest tests/test_offline_inference.py`.
 """
 import os
 
-import pytest
 import vllm  # noqa: F401
 
 from tests.conftest import VllmRunner
@@ -47,8 +46,6 @@ def test_models_distributed_QwQ():
         vllm_model.generate_greedy(example_prompts, max_tokens)
 
 
-@pytest.mark.skipif(os.getenv("VLLM_USE_V1") == "1",
-                    reason="deepseek v2 lite is not supported on v1")
 def test_models_distributed_DeepSeek():
     example_prompts = [
         "vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs.",

vllm_ascend/attention/mla_v1.py

@@ -239,10 +239,8 @@ class AscendMLAMetadataBuilder:
         # it blocks on all previous kernels.
         device = self.runner.device
 
-        block_table = self.runner.input_batch.block_table[0].get_device_tensor(
-        )
-        block_table[:num_reqs, :self.runner.max_num_blocks_per_req] = (
-            block_table[:num_reqs])
+        block_table = (self.runner.input_batch.block_table[0].
+                       get_device_tensor()[:num_reqs])
         slot_mapping = self.runner.slot_mapping_cpu[:num_actual_tokens].to(
             device, non_blocking=True)
         input_positions = self.runner.positions_cpu[:num_actual_tokens].to(