🚨 [v5] Toggle the serialization format in processors

[zucchini-nlp]

What does this PR do?

As per title, in v5 we will start saving in the new nested format. The loading is still supported though for old-format configs

Also, we don't need an optional_attributes field that does not do what it is called. Since it is only chat template, let's just manually pop and set it. For audio tokenizers, it'd be nice to refactor a bit, I think we could try to move it in self.attributes

Fixes #40447 by deleting ambiguously named optional_attributes

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: align, auto, chinese_clip, clipseg, flava, granite_speech, layoutlmv2, layoutlmv3, markuplm, mgp_str, oneformer, owlvit, speech_to_text, speecht5, vision_text_dual_encoder, wav2vec2