[ernie-image] use concrete Mistral3Model / Ministral3ForCausalLM types

[akshan-main]

What does this PR do?

Follow-up to #13663 (per @asomoza's review there): replace the AutoModel / AutoModelForCausalLM references for text_encoder and pe with the concrete Mistral3Model / Ministral3ForCausalLM so the loaded checkpoint matches the declared types.

To avoid breaking installs with older transformers (Ministral3ForCausalLM is only in 5.0+, Mistral3Model is in 4.50+):

• Standard: TYPE_CHECKING + string annotations on __init__ so the imports never run at module load. _get_signature_types uses get_type_hints and falls back gracefully when a name can't be resolved.
• Modular encoders.py: try/except import that aliases the concrete class when available and falls back to AutoModel / AutoModelForCausalLM otherwise. The ComponentSpec then references the alias.

So users on recent transformers get the silenced warning; users on older transformers keep working as before.

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? Address ernie-image review findings #13577 #13663
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

Who can review?

@yiyixuxu @asomoza

[asomoza]

thanks @akshan-main for what understand, this is not the best fix, since with the fallback (as it was before) it will still not work because the models are not present in transformers. So probably instead of the fallback we should raise an error but the version depends if they use or not the PE @yiyixuxu WDYT?

[yiyixuxu]

ohh I just thought it was an issue with CI on our end, not something you need to fix
can you just keep the change (not the fix), so that we can see if the CI fails here and look into it

[akshan-main]

@yiyixuxu kept just the type change

[yiyixuxu]

can you try update the minimum heree https://github.com/huggingface/diffusers/blob/main/setup.py#L143
not sure if it'd fix this

cc @sayakpaul @DN6
we also run into this in #13551

[akshan-main]

@yiyixuxu done

[sayakpaul]

Hmm, I think Ministral3ForCausalLM should be guarded with something like:

if is_transformers_version("<", "5.0.0"):
   raise

This way, I don't think we have to update the setup.py (I don't mind the change, either as we made Diffusers compatible with Transformers v5 a while back). In our CI, we install transformers from main.

[akshan-main]

@sayakpaul @yiyixuxu applied it, lmk if I need to revert