Fix TypeError in LongCatImageEditPipeline truncation warning

[Ricardo-M-L]

What this PR does

LongCatImageEditPipeline._encode_prompt calls len() twice on an int when building its truncation warning message:

https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/longcat_image/pipeline_longcat_image_edit.py#L284-L289

if len(all_tokens) > self.tokenizer_max_length:
    logger.warning(
        "Your input was truncated because \`max_sequence_length\` is set to "
        f" {self.tokenizer_max_length} input token nums : {len(len(all_tokens))}"
    )
    all_tokens = all_tokens[: self.tokenizer_max_length]

len(all_tokens) already returns an int, so the outer len() raises TypeError: object of type 'int' has no len() — inside the f-string formatting — every time the branch fires.

Why this is a real bug

The warning only executes when len(all_tokens) > self.tokenizer_max_length (default 512). That is the exact scenario the warning is supposed to inform the user about. Instead of informing them, the pipeline crashes with a TypeError during prompt encoding.

Minimal repro:

pipe = LongCatImageEditPipeline.from_pretrained(...)
pipe(prompt="a very long prompt ..." * 200, image=image)
# TypeError: object of type 'int' has no len()

The sibling LongCatImagePipeline._encode_prompt has the correct form at line 291 — this is a local typo in the edit pipeline only.

Fix

Drop the extraneous len() call to match the sibling pipeline:

-                f" {self.tokenizer_max_length} input token nums : {len(len(all_tokens))}"
+                f" {self.tokenizer_max_length} input token nums : {len(all_tokens)}"

Before submitting

• Did you read the contributor guideline?
• Was this discussed/approved via a Github issue or the forum? N/A — single-character typo fix.
• Did you make sure to update the documentation with your changes? N/A.
• Did you write any new necessary tests? N/A — fixes a log-message crash; no behavior change.

Who can review?

@yiyixuxu @sayakpaul