[Flux] Add advanced training script + support textual inversion inference

[linoytsaban]

This PR adds an advanced version of the dreambooth lora flux script, with additional accompanying update to FluxPipline:

• adds an advanced script with pivotal tuning feature for CLIP & T5 encoder

• • --train_text_encoder_ti (enables CLIP pivotal tuning)
• • --enable_t5_ti adds T5 to the mix
• • --train_text_encoder_ti_frac the portion of epochs to train embeddings on, when using just clip
• • --train_transformer_frac the portion of epochs to train the transformer on, train_transformer_frac==0 will trigger a "pure_textual_inversion" train run (= "classical" textual inversion, no optimization of transformer lora layers)
• • --initializer_token the token to init the textual inversion embeddings with instead of randomly (random by default).

• --lora_blocks the blocks\layers to apply lora training on.

• modifies Flux Pipeline (and related pipelines) to allow for textual inversion inference (that is also required by Loras trained with pivotal tuning)

motivation: to allow for fast iterations & experimental features. I think it'd be good to fork the canonical script and bring the above changes into the advanced folder in a similar manner to what we have for SDXL.

cc @apolinario

[linoytsaban]

@sayakpaul @apolinario wdyt about me closing #9160 and moving the changes here as I suggest above?

[sayakpaul]

I would prefer #9160 as it helps to review the changes to the canonical script in isolation. Would that work for you?

[sayakpaul]

@linoytsaban let me know when you need another review.

[sayakpaul]

Thanks!

[sayakpaul]

@yiyixuxu could you also give the changes in the pipelines a look? It's just about adding TextualInversionMixin so that we can enable pivotal tuning on Flux.

[yiyixuxu]

thanks!

[apolinario]

Let's go! 🚀