[Inference/Kernel] Add Paged Decoding kernel, sequence split within the same thread block (#5531)

* feat flash decoding for paged attention

* refactor flashdecodingattention

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

colossalai/inference/modeling/models/nopadding_llama.py

@@ -437,6 +437,19 @@ class NopadLlamaAttention(LlamaAttention):
                     block_tables,
                     high_precision,
                 )
+                # inference_ops.flash_decoding_attention(
+                #     attn_output,
+                #     query_states,
+                #     k_cache,
+                #     v_cache,
+                #     sequence_lengths,
+                #     block_tables,
+                #     block_size,
+                #     kv_seq_len,
+                #     fd_inter_tensor.mid_output,
+                #     fd_inter_tensor.mid_output_lse,
+                #     sm_scale,
+                # )
             else:
                 if is_verifier:
                     rotary_embedding(query_states, key_states, cos_sin[0], cos_sin[1])