hotfix: revert sampler CUDA Graph

[zhyncs]

Motivation

Modifications

Checklist

• Format your code according to the Contributor Guide.
• Add unit tests as outlined in the Contributor Guide.
• Update documentation as needed, including docstrings or example tutorials.

[zhyncs]

1. sampler CUDA Graph
2. bench latency
3. new release v0.2.14.post1

[zhyncs]

python3 -m sglang.bench_latency --model meta-llama/Meta-Llama-3.1-8B-Instruct  --correct --output-len 16 --trust-remote-code
Init nccl begin.
Load weight begin. avail mem=78.73 GB
INFO 08-28 11:11:49 weight_utils.py:236] Using model weights format ['*.safetensors']
Loading safetensors checkpoint shards:   0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards:  25% Completed | 1/4 [00:00<00:00,  3.92it/s]
Loading safetensors checkpoint shards:  50% Completed | 2/4 [00:01<00:01,  1.48it/s]
Loading safetensors checkpoint shards:  75% Completed | 3/4 [00:02<00:00,  1.07it/s]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:03<00:00,  1.03s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:03<00:00,  1.10it/s]

Load weight end. type=LlamaForCausalLM, dtype=torch.bfloat16, avail mem=63.65 GB
Memory pool end. avail mem=8.39 GB
Capture cuda graph begin. This can take up to several minutes.
max_total_num_tokens=443991
input_ids=[[128000, 791, 6864, 315, 9822, 374], [128000, 791, 6864, 315, 279, 3723, 17262, 316, 374], [128000, 15724, 374, 264, 40798, 1938, 323, 358, 1093]]
prefill logits (first half) tensor([[ 2.4688,  1.4844,  0.5117,  ..., -2.5000, -2.5000, -2.5000],
        [ 2.4688,  1.4844,  0.5117,  ..., -2.5000, -2.5000, -2.5000],
        [ 2.0312,  2.1562,  2.2344,  ..., -6.6875, -6.6875, -6.6875]],
       device='cuda:0')
prefill logits (final) tensor([[ 5.1875,  2.0156,  1.0469,  ..., -5.2188, -5.2188, -5.2188],
        [ 4.9688,  3.5156,  2.2812,  ..., -3.7812, -3.7812, -3.7812],
        [10.2500,  3.1719,  2.1719,  ..., -2.6875, -2.6875, -2.6875]],
       device='cuda:0')
<|begin_of_text|>The capital of France is a city of romance, art, fashion, and cuisine. Paris is a must-
<|begin_of_text|>The capital of the United Kindom is London. London is a global city and a major financial center. It is the most
<|begin_of_text|>Today is a sunny day and I like to take a walk in the park. I put on my sunglasses and my favorite hat

python3 scripts/playground/reference_hf.py --model  meta-llama/Meta-Llama-3.1-8B-InstructLoading checkpoint shards: 100%|██████████████| 4/4 [00:30<00:00,  7.69s/it]
/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:567: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0.6` -- this flag is only used in sample-based generationmodes. You should set `do_sample=True` or unset `temperature`.
  warnings.warn(
/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:572: UserWarning: `do_sample` is set to `False`. However, `top_p` is set to `0.9` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `top_p`.
  warnings.warn(
The attention mask and the pad token id were not set. As a consequence, youmay observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.
The attention mask is not set and cannot be inferred from input because padtoken is same as eos token. As a consequence, you may observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
prefill logits tensor([ 5.2227,  2.0332,  1.0615,  ..., -5.2305, -5.2305, -5.2305],
       device='cuda:0')
<|begin_of_text|>The capital of France is a city of romance, art, fashion, and cuisine. Paris is a must
The attention mask and the pad token id were not set. As a consequence, youmay observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.
prefill logits tensor([ 4.9688,  3.5391,  2.2910,  ..., -3.7285, -3.7285, -3.7285],
       device='cuda:0')
<|begin_of_text|>The capital of the United Kindom is London. London is a global city and a major financial center. It is the
The attention mask and the pad token id were not set. As a consequence, youmay observe unexpected behavior. Please pass your input's `attention_mask` to obtain reliable results.
Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.
prefill logits tensor([10.2422,  3.1367,  2.1719,  ..., -2.6777, -2.6777, -2.6777],
       device='cuda:0')
<|begin_of_text|>Today is a sunny day and I like to take a walk in the park. I put on my sunglasses and my favorite