Add error-checking branch to catch lists of non-tensors being passed as ip_adapter_image_embeds

[beyarkay]

What does this PR do?

This PR just adds an extra branch to the parameter checking (in self.check_inputs) which makes sure that the value being passed in as ip_adapter_image_embeds is indeed a list of items which have a ndim attribute. I mistakenly passed in a list of strings and got hit with:

  File "/Users/brk/projects/quick-ideas/.venv/lib/python3.10/site-packages/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py", line 681, in check_inputs
    elif ip_adapter_image_embeds[0].ndim not in [3, 4]:
AttributeError: 'str' object has no attribute 'ndim'

Which took looking into the source code to figure out. After this commit, the error is:

  File "/Users/brk/projects/quick-ideas/.venv/lib/python3.10/site-packages/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py", line 682, in check_inputs
    raise ValueError(
ValueError: `ip_adapter_image_embeds` must be a list of tensors, but is a list of <class 'str'>

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline?
• Did you read our philosophy doc (important for complex PRs)?
• Was this discussed/approved via a GitHub issue or the forum? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

(tagging because pipelines: @sayakpaul @yiyixuxu)

[sayakpaul]

@fabiorigano WDYT about this?

[fabiorigano]

hi everyone, thanks, @sayakpaul, for adding me here
I suggested a change to the current solution. All pipelines should be updated, right?

[fabiorigano]

Suggested change

	elif not hasattr(ip_adapter_image_embeds[0], "ndim"):
	elif not isinstance(ip_adapter_image_embeds[0], torch.Tensor):

Both torch.Tensor and numpy.ndarray have the ndim attribute. If an user passes a list of numpy.ndarray objects as ip_adapter_image_embeds, he will get an error in the prepare_ip_adapter_image_embeds method, because this object does not have a chunk method.

[beyarkay]

Sounds good to me 👍