Remove output_type="latent" from modular decoders

[dhvani7014]

What does this PR do?

Fixes #15071

Removes the output_type == "latent" branches from the modular decode blocks for flux, sd3, sdxl, cosmos and echo. Latents are available via output="latents" instead. ltx2 is left for #14612. Will rebase after #14737 merges.

Self-review notes

• Removed the output_type="latent" branches from the flux, sd3, sdxl, cosmos and echo modular decoders.
• Passing output_type="latent" now raises a ValueError pointing to output="latents", same as minimax_h3. Happy to switch to a deprecation warning if you prefer.
• Dropped the unused output_type input from EchoAudioDecoderStep.
• Tests that used "latent" now read latents/audio_latents from the output or use output_type="pt". The echo audio stats test checks _denormalize_audio_latents directly, since the decoder can no longer return latents.
• Updated the echo docs line that described the latent output.
• Ran make style, make fix-copies, and pytest tests/modular_pipelines/echo tests/modular_pipelines/cosmos: 93 passed, 25 skipped.

Before submitting

• Did you use an AI agent (Claude Code, Codex, Cursor, etc.) to help with this PR? If so:
  • Did you read the Coding with AI agents guide?
  • Did you run the self-review skill on the diff?
  • Did you share the final self-review notes in the PR description or a comment?
• 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. Follow-up to #14737: drop output_type="latent" from modular decode blocks #15071
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?
• Are you the author (or part of the team) of the model/pipeline (only applicable for model/pipeline related PRs)?

Who can review?

@yiyixuxu