[Feature] Unify fp8 block_wise quant ops (#5991)

* quant stash

* blockwise_quant

* precommit

* rm tensor.cut

* tp ok

* add swiglu

* rm outdate code

* fix activate ut

* change baseline

* fix baseline error

fastdeploy/model_executor/layers/quantization/block_wise_fp8.py

@@ -18,7 +18,6 @@ from typing import Optional
 
 import paddle
 
-import fastdeploy
 from fastdeploy import envs
 from fastdeploy.model_executor.layers.linear import (
     MergedColumnParallelLinear,
@@ -264,9 +263,10 @@ class BlockWiseFP8LinearMethod(QuantMethodBase):
         layer.weight_scale_inv.set_value(weight_scale)
 
     def apply(self, layer, x):
-        x, x_scale_tensor = fastdeploy.model_executor.ops.gpu.per_token_quant_padding(
-            x, self.quant_config.weight_block_size[0]
+        x, x_scale_tensor = paddle.incubate.nn.functional.fp8_quant_blockwise(
+            x, using_pow2_scale=False, output_scale_transpose=True
         )
+        x_scale_tensor = x_scale_tensor.T
         linear_out = paddle.empty((x.shape[0], layer.output_size), dtype=paddle.bfloat16)
         linear_out = deep_gemm_fp8_fp8_bf16_nt(
             x, x_scale_tensor, layer.weight, layer.weight_scale_inv, linear_out, layer.output_size