LTX2Pipeline: dynamic timestep shift `mu` is constant, ignores the actual sequence length

In LTX2Pipeline.__call__ (src/diffusers/pipelines/ltx2/pipeline_ltx2.py), mu is computed with max_image_seq_len passed as the first argument of calculate_shift — the image_seq_len parameter:

mu = calculate_shift(
    self.scheduler.config.get("max_image_seq_len", 4096),   # <-- this is image_seq_len; should be the sample's real sequence length
    self.scheduler.config.get("base_image_seq_len", 1024),
    self.scheduler.config.get("max_image_seq_len", 4096),
    self.scheduler.config.get("base_shift", 0.95),
    self.scheduler.config.get("max_shift", 2.05),
)

calculate_shift(image_seq_len, base, max, base_shift, max_shift) returns image_seq_len * m + b. With image_seq_len == max_seq_len, that reduces to max_shift, so mu is constant regardless of resolution or frame count — use_dynamic_shifting=True has no effect.

image_seq_len should be the sample's packed sequence length, latent_num_frames * latent_height * latent_width — which is already computed a few lines above (and even present as the commented line # video_sequence_length = latent_num_frames * latent_height * latent_width). This matches the Flux/SD3 pipelines and the LTX reference (math.prod(latent.shape[2:])).

Fix: pass latent_num_frames * latent_height * latent_width as the first argument.

Comienza en src/diffusers/pipelines/ltx2/pipeline_ltx2.py, dentro de LTX2Pipeline.call, y lee los cálculos de las dimensiones latentes cercanos y la llamada a calculate_shift. Confirma cómo se deriva la longitud de la secuencia empaquetada y compara la lógica análoga de la pipeline Flux/SD3. Se considera terminado cuando el desplazamiento dinámico usa la longitud real de la secuencia latente y varía según la resolución o el número de frames.