model: support GLM4V vision encoder (#18042)

* convert ok

* no deepstack

* less new tensors

* cgraph ok

* add mrope for text model

* faster patch merger

* add GGML_ROPE_TYPE_MRNORM

* add support for metal

* move glm4v do dedicated graph

* convert: add norm_embd

* clip: add debugging fn

* working correctly

* fix style

* use bicubic

* fix mrope metal

* improve cpu

* convert to neox ordering on conversion

* revert backend changes

* force stop if using old weight

* support moe variant

* fix conversion

* fix convert (2)

* Update tools/mtmd/clip-graph.h

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* process mrope_section on TextModel base class

* resolve conflict merge

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

tools/mtmd/clip-graph.h

@@ -9,6 +9,8 @@
 #include <vector>
 #include <functional>
 
+#define DEFAULT_INTERPOLATION_MODE (GGML_SCALE_MODE_BILINEAR | GGML_SCALE_FLAG_ANTIALIAS)
+
 struct clip_graph {
     const clip_model & model;
     const clip_hparams & hparams;
@@ -49,7 +51,7 @@ struct clip_graph {
     void cb(ggml_tensor * cur0, const char * name, int il) const;
 
     // siglip2 naflex
-    ggml_tensor * resize_position_embeddings();
+    ggml_tensor * resize_position_embeddings(uint32_t interpolation_mode = DEFAULT_INTERPOLATION_MODE);
 
     // build vision transformer (ViT) cgraph
     // this function should cover most of the models