[FIX] The diffusion pipeline returned dtype error

Due to `torch_dtype = torch.float16` the pipeline returned error `RuntimeError: mat1 and mat2 must have the same dtype`
Cause the dtype of tensor and models layer dtype didn't match. [link_to_issue](https://discuss.pytorch.org/t/runtimeerror-mat1-and-mat2-must-have-the-same-dtype/166759)

README.md

@@ -38,7 +38,7 @@ pipe = StableDiffusionControlNetImg2ImgPipeline.from_pretrained(
     "runwayml/stable-diffusion-v1-5",
     controlnet=controlnet,
     safety_checker=None,
-    torch_dtype=torch.float16
+    torch_dtype=torch.float32
 )
 
 pipe.enable_xformers_memory_efficient_attention()