Adding `safetensors` variant of these models

This PR adds a `safetensors` variant of these models.

This is essentially identical to the PyTorch state dict file, but safe in the sense that no arbitrary code can be put into it.

These files also happen to load much faster than their pytorch counterpart:
https://colab.research.google.com/github/huggingface/notebooks/blob/main/safetensors_doc/en/speed.ipynb

also hi nubby lol <3

(what, did you think i'd just use their tool instead of hacking it up myself?)

WD1-4-kl-f8-anime2-bless08.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0735d465545f84b3368ff17d6517fe3a9c0682074f246a81b978903e5e1fdff7
+size 334641012

WD1-4-kl-f8-anime2-bless09.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c32b727313df0717f05b45e133574f9db75a7854d572392bd74de1c3fb3df79f
+size 334641012

WD1-4-kl-f8-anime2-bless1-05.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c66161002b8e024d8c40efc6d183714fa904e81aef794223850a678205309051
+size 334641012

WD1-4-kl-f8-anime2-bless1-1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1f7b2a64fdc8d8d62c2497e99c07e9f0b755b8a1530072ed5d5d8e10f4e41f4
+size 334641012