[KV-Cache] Make k_scale, v_scale as attributes of self_attn using HFCache

[horheynm]

FIX #132
Automatically set k/v_scales a part of self_attn module's attribute.

Before:
Compute input, weight, output quantization params, create it as an attribute of the leaf module (ex. self_attn.k_proj.output_scale)
Then, copy the attribute to the self_attn layer (ex. self_attn.k_proj.output_scale -> self_attn.k_scale) and delete the attr from the child.

After
Use HF's cache object, when self_attn forward is called, use the cache to compute the k_scale and v_scale of the kv cache, wrap the forward call to automatically populate the scales to self_attn.

Note:
Previously, the scales were computed from wrap_module_forward_quantized, which does out = output_fq(weight_fq(input_fq(x))where .*_fq is fake quantize

Now the cache computes the k/v_scales of the kv cache, not the output activations

[mgoin]

I tested on a few different models and it works well for producing checkpoints that work with vLLM, thanks George!