feat(api): report model memory placement in memory endpoint

Add a model section to /sdapi/v1/memory with loaded-model bytes summed
per pipeline component and device, so clients can tell resident weights
from offloaded ones and loop-critical components from edge ones.

- walk components over parameters and buffers, dedupe shared storages,
  key by component name then device type
- read the raw model slot so a memory poll never triggers a model load
- section is exception-isolated like ram and cuda; reports an error
  string if the walk races a reload

modules/api/models.py

@@ -552,6 +552,7 @@ class ResEmbeddings(BaseModel):
 class ResMemory(BaseModel):
     ram: dict = Field(title="RAM", description="System memory stats")
     cuda: dict = Field(title="CUDA", description="nVidia CUDA memory stats")
+    model: dict = Field(default={}, title="Model", description="Loaded model bytes per component and device")
 
 class ResScripts(BaseModel):
     txt2img: list[str] = Field(title="Txt2img", description="Titles of scripts (txt2img)")