从 0.3.51 开始，执行 python main.py --lowvram 显存无法成功利用，更新到 0.3.56 也依然如此

### Custom Node Testing

- [x] I have tried disabling custom nodes and the issue persists (see [how to disable custom nodes](https://docs.comfy.org/troubleshooting/custom-node-issues#step-1%3A-test-with-all-custom-nodes-disabled) if you need help)

### Expected Behavior

Using split attention in VAE
Using split attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
Using scaled fp8: fp8 matrix mult: False, scale input: False
Requested to load QwenImageTEModel_
loaded completely 9.5367431640625e+25 7263.013442993164 True
CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
FETCH ComfyRegistry Data: 75/96
FETCH ComfyRegistry Data: 80/96
FETCH ComfyRegistry Data: 85/96
model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
model_type FLUX
FETCH ComfyRegistry Data: 90/96
FETCH ComfyRegistry Data: 95/96
Requested to load QwenImage
FETCH ComfyRegistry Data [DONE]
[ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json [DONE]
[ComfyUI-Manager] All startup tasks have been completed.
loaded partially 11445.99951171875 11445.237854003906 0
100%|███████████████████████████████████████████████████████████████████████████████████| 20/20 [01:27<00:00, 4.38s/it]
Requested to load WanVAE
loaded completely 4917.515625 242.02829551696777 True
Prompt executed in 127.56 seconds
got prompt
Requested to load QwenImage
loaded partially 9779.47998046875 9778.345275878906 0
100%|███████████████████████████████████████████████████████████████████████████████████| 20/20 [01:24<00:00, 4.22s/it]
Requested to load WanVAE

### Actual Behavior

To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/96
got prompt
Using split attention in VAE
Using split attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
Using scaled fp8: fp8 matrix mult: False, scale input: False
Requested to load QwenImageTEModel_
loaded completely 9.5367431640625e+25 7909.737449645996 True
CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
FETCH ComfyRegistry Data: 10/96
FETCH ComfyRegistry Data: 15/96
FETCH ComfyRegistry Data: 20/96
FETCH ComfyRegistry Data: 25/96
model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
model_type FLUX
FETCH ComfyRegistry Data: 30/96
Killed

### Steps to Reproduce

git checkout 0.3.51

python main.py --lowvram

### Debug Logs

```powershell
To see the GUI go to: http://127.0.0.1:8188
FETCH ComfyRegistry Data: 5/96
got prompt
Using split attention in VAE
Using split attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.float16
Using scaled fp8: fp8 matrix mult: False, scale input: False
Requested to load QwenImageTEModel_
loaded completely 9.5367431640625e+25 7909.737449645996 True
CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16
FETCH ComfyRegistry Data: 10/96
FETCH ComfyRegistry Data: 15/96
FETCH ComfyRegistry Data: 20/96
FETCH ComfyRegistry Data: 25/96
model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
model_type FLUX
FETCH ComfyRegistry Data: 30/96
Killed
```

### Other

WSL environment detected.
=====================
HSA System Attributes
=====================
Runtime Version: 1.1
Runtime Ext Version: 1.7
System Timestamp Freq.: 1000.000000MHz
Sig. Max Wait Duration: 18446744073709551615 (0xFFFFFFFFFFFFFFFF) (timestamp count)
Machine Model: LARGE
System Endianness: LITTLE
Mwaitx: DISABLED
XNACK enabled: NO
DMAbuf Support: YES
VMM Support: YES

==========
HSA Agents
==========

*******
Agent 2
*******
Name: gfx1201
Marketing Name: AMD Radeon RX 9070 XT
Vendor Name: AMD
Feature: KERNEL_DISPATCH
Profile: BASE_PROFILE
Float Round Mode: NEAR
Max Queue Number: 128(0x80)
Queue Min Size: 64(0x40)
Queue Max Size: 131072(0x20000)
Queue Type: MULTI
Node: 1
Device Type: GPU
Cache Info:
L1: 32(0x20) KB
L3: 65536(0x10000) KB
Chip ID: 30032(0x7550)
Cacheline Size: 64(0x40)
Max Clock Freq. (MHz): 2520
Internal Node ID: 1
Compute Unit: 64
SIMDs per CU: 2
Shader Engines: 4
Shader Arrs. per Eng.: 2
Coherent Host Access: FALSE
Memory Properties:
Features: KERNEL_DISPATCH
Fast F16 Operation: TRUE
Wavefront Size: 32(0x20)
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Max Waves Per CU: 32(0x20)
Max Work-item Per CU: 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
Max fbarriers/Workgrp: 32
Packet Processor uCode:: 48
SDMA engine uCode:: 0
IOMMU Support:: None
Pool Info:
Pool 1
Segment: GLOBAL; FLAGS: COARSE GRAINED
Size: 16695296(0xfec000) KB
Allocatable: TRUE
Alloc Granule: 4KB
Alloc Recommended Granule:2048KB
Alloc Alignment: 4KB
Accessible by all: FALSE
Pool 2
Segment: GROUP
Size: 64(0x40) KB
Allocatable: FALSE
Alloc Granule: 0KB
Alloc Recommended Granule:0KB
Alloc Alignment: 0KB
Accessible by all: FALSE
ISA Info:
ISA 1
Name: amdgcn-amd-amdhsa--gfx1201
Machine Models: HSA_MACHINE_MODEL_LARGE
Profiles: HSA_PROFILE_BASE
Default Rounding Mode: NEAR
Default Rounding Mode: NEAR
Fast f16: TRUE
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
FBarrier Max Size: 32
ISA 2
Name: amdgcn-amd-amdhsa--gfx12-generic
Machine Models: HSA_MACHINE_MODEL_LARGE
Profiles: HSA_PROFILE_BASE
Default Rounding Mode: NEAR
Default Rounding Mode: NEAR
Fast f16: TRUE
Workgroup Max Size: 1024(0x400)
Workgroup Max Size per Dimension:
x 1024(0x400)
y 1024(0x400)
z 1024(0x400)
Grid Max Size: 4294967295(0xffffffff)
Grid Max Size per Dimension:
x 4294967295(0xffffffff)
y 4294967295(0xffffffff)
z 4294967295(0xffffffff)
FBarrier Max Size: 32
*** Done ***