SDXL TensorRT + IP Adapter

### Model/Pipeline/Scheduler description

Dear authors,

I was wondering if there will be a pipeline that support SDXL + IP Adapter + TensorRT in near future? In the original implementation of TensorRT, they've implemented the SDXL + TensorRT but lacking the IP Adapter/ T2I Adapter...

### Open source status

- [X] The model implementation is available.
- [X] The model weights are available (Only relevant if addition is not a scheduler).

### Provide useful links for the implementation

Official SDXL TensorRT implementation:
https://github.com/NVIDIA/TensorRT/blob/release/10.1/demo/Diffusion/demo_txt2img_xl.py

Code should be added in the following file:
https://github.com/NVIDIA/TensorRT/blob/release/10.1/demo/Diffusion/models.py

```
class IPAdapterLoader(IPAdapterMixin):
def __init__(self, paths):
pass
```

Comienza comparando la implementación enlazada de TensorRT SDXL con la ubicación solicitada de models.py y el stub de IPAdapterLoader. Determina cómo debería conectarse la compatibilidad con IP Adapter a la canalización de SDXL TensorRT y, a continuación, verifica que la canalización resultante admita la combinación solicitada de SDXL, IP Adapter y TensorRT.

Escrito por el modelo de indexación a partir del texto del issue.