model: support youtu-vl model (#18479)

* Support Youtu-VL Model

* merge code

* fix bug

* revert qwen2 code & support rsplit in minja.hpp

* update warm info

* fix annotation

* u

* revert minja.hpp

* fix

* Do not write routed_scaling_factor to gguf when routed_scaling_factor is None

* fix expert_weights_scale

* LGTM after whitespace fixes

* fix

* fix

* fix

* layers to layer_index

* enum fix

---------

Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

gguf-py/gguf/constants.py

@@ -294,7 +294,9 @@ class Keys:
         USE_GELU            = "clip.use_gelu"
         USE_SILU            = "clip.use_silu"
         N_WA_PATTERN        = "clip.vision.n_wa_pattern" # used by qwen2.5vl
+        WA_LAYER_INDEXES    = "clip.vision.wa_layer_indexes" # used by youtuvl
         IS_DEEPSTACK_LAYERS = "clip.vision.is_deepstack_layers"
+        WINDOW_SIZE         = "clip.vision.window_size"
 
         class Attention:
             HEAD_COUNT      = "clip.vision.attention.head_count"
@@ -3494,6 +3496,7 @@ class VisionProjectorType:
     LFM2A = "lfm2a" # audio
     MUSIC_FLAMINGO = "musicflamingo" # audio
     GLM4V = "glm4v"
+    YOUTUVL = "youtuvl"
 
 
 # Items here are (block size, type size)