[Refactor] Integrated some lightllm kernels into token-attention (#4946)

* add some req for inference

* clean codes

* add codes

* add some lightllm deps

* clean codes

* hello

* delete rms files

* add some comments

* add comments

* add doc

* add lightllm deps

* add lightllm cahtglm2 kernels

* add lightllm cahtglm2 kernels

* replace rotary embedding with lightllm kernel

* add some commnets

* add some comments

* add some comments

* add

* replace fwd kernel att1

* fix a arg

* add

* add

* fix token attention

* add some comments

* clean codes

* modify comments

* fix readme

* fix bug

* fix bug

---------

Co-authored-by: cuiqing.li <lixx336@gmail.com>
Co-authored-by: CjhHa1 <cjh18671720497@outlook.com>

colossalai/kernel/triton/self_attention_nofusion.py

@@ -12,6 +12,7 @@ if HAS_TRITON:
     from .qkv_matmul_kernel import qkv_gemm_4d_kernel
     from .softmax import softmax_kernel
 
+    # adpeted from https://github.com/microsoft/DeepSpeed/blob/master/deepspeed/ops/transformer/inference/triton/triton_matmul_kernel.py#L312
     def self_attention_forward_without_fusion(
         q: torch.Tensor, k: torch.Tensor, v: torch.Tensor, input_mask: torch.Tensor, scale: float
     ):
@@ -141,6 +142,7 @@ if HAS_TRITON:
         )
         return output.view(batches, -1, d_model)
 
+    # modified from https://github.com/microsoft/DeepSpeed/blob/master/deepspeed/ops/transformer/inference/triton/attention.py#L212
     def self_attention_compute_using_triton(
         qkv, input_mask, layer_past, alibi, scale, head_size, triangular=False, use_flash=False
     ):