fix wan i2v num_videos batching, vace latent trim, config scale factors, processor config, modular dtype

[akshan-main]

What does this PR do?

Part of #13578 (wan review). Same families of issues as the qwenimage review:

• WanImageToVideoPipeline conditioning batching: image_embeds was repeated by batch_size only (not num_videos_per_prompt) while the latents reach batch_size * num_videos_per_prompt, so num_videos_per_prompt > 1 failed the transformer concat; and list image inputs (a documented PipelineImageInput type) were rejected in check_inputs. Now check_inputs accepts lists, and both image_embeds and the conditioning latents expand from a single image or a per-prompt list up to the effective batch. The first+last-frame path keeps its two-frame layout.
• WanVACEPipeline(output_type="latent"): the decode path trims the prepended reference latents but the latent path returned them untrimmed, so latent output carried extra frames. The trim now runs before the branch.
• WanVACEPipeline, WanVideoToVideoPipeline, and the modular Wan pipeline derived scale factors from len(vae.temperal_downsample) and a hardcoded spatial 8 instead of vae.config.scale_factor_spatial / scale_factor_temporal. Now read from config, matching the base WanPipeline.
• WanAnimateImageProcessor.__init__ accepted vae_scale_factor / vae_latent_channels / resample etc. but called super().__init__() with no args, so the base re-registered defaults over the subclass config. Now forwards them.
• Modular WanTextInputStep derived block_state.dtype from prompt_embeds.dtype; the denoise blocks cast scheduler timesteps to it, losing precision on bf16/fp16 paths. Now uses transformer.dtype, matching the comment.

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? Discussed on Slack with @yiyixuxu
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

Who can review?

@yiyixuxu

[github-actions]

Hi @akshan-main, thanks for the PR! It does not appear to link an issue it fixes. If this PR addresses an existing issue, please add a closing keyword (e.g. Fixes #1234) to the PR description so the issue is linked. See the contribution guide for more details. If this PR intentionally does not fix a tracked issue, a maintainer can add the no-issue-needed label to silence this reminder.