[SOT] Mark dynamic dims by type annotations (#2771)

* [SOT] Mark dynamic dims by type annotations

* fix conflict of forward_meta

* mark more attn backend

* fix missing annotated and add env SOT_SPECIALIZED_DIM_NUMBERS

* auto infer implicit 0 dim dynamic dim

* revert manual marked dims

* revert missing update

* auto infer can use unsafe code in warmup stage

* check -> type_match

* fix codestyle

* restore blank line

* empty commit

* add need_warmup nonlocal;

* add doc for resolver

* add missing type hints

* unquote "ForwardMeta"

fastdeploy/model_executor/layers/attention/__init__.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 
 from .append_attn_backend import AppendAttentionBackend
+from .attention import Attention
 from .attention_selecter import get_attention_backend
 from .base_attention_backend import AttentionBackend
 from .block_multihead_attn_backend import BlockAttentionBackend
@@ -32,4 +33,5 @@ __all__ = [
     "FlashAttentionBackend",
     "IluvatarAttnBackend",
     "BlockAttentionBackend",
+    "Attention",
 ]