convert : support rope_scaling type and rope_type

[CISC]

At some point transformers renamed rope_scaling type to rope_type, so support both.

[ngxson]

The same code is copied in multiple places, so I think it's better to group them into a new function like self.set_rope_config()

Edit: or we can extend self.set_gguf_parameters() to support this

Btw, which model(s) you have been testing with?

[CISC]

The problem with extending the base set_gguf_parameters method is that you will have to merge all special cases, like f.ex. DeepseekV2Model, it can get messy.

I tested with a few models I still had the original files for, not every single one I touched, but I'm fairly sure I didn't break anything. :)

[ngxson]

I still think having this as a dedicated function like self.set_rope_config() will make it easier to maintain. We can optionally check if rope_scaling["factor"] has a good value (i.e. non-null), but it's up to you anyway.

[CISC]

I absolutely agree, but I worry about the special cases. :)

I'll see what I can do...

[CISC]

I will merge this as-is for now and make a new PR later.

Deduplication of the rope code requires careful thought (I think we also can deduplicate the llama3 rope_freqs calculations).