Port of Qwen3-VL support from mainline (#883)

* Port of Qwen3-VL for latest ik_llama.cpp

- convert_hf_to_gguf.py - Not touched, use llama.cpp to convert model instead
- sysl and metal support for imrope not added
- Vulkan support for imrope not tested
- Code not tested

* Bugfix n_embd was declared multiple times

https://github.com/ikawrakow/ik_llama.cpp/pull/883#issuecomment-3471179655

* Fix n_embd issue with qwen3vl

* model.output tensor not required

https://github.com/ikawrakow/ik_llama.cpp/pull/883#discussion_r2480388389

* Improved logic for qkv combined tensors

59ceaf8fcb (r2480395800)
59ceaf8fcb (r2480398187)

* Fix n_embd for merge_qkv() + cleaner code

https://github.com/ikawrakow/ik_llama.cpp/pull/883#discussion_r2481227395

* Revert TENSOR_NOT_REQUIRED

src/llama-model.h

@@ -24,22 +24,39 @@ enum e_model {
     MODEL_80M,
     MODEL_109M,
     MODEL_137M,
+    MODEL_140M,
     MODEL_160M,
+    MODEL_190M,
     MODEL_220M,
     MODEL_250M,
+    MODEL_256M,
     MODEL_270M,
     MODEL_335M,
+    MODEL_350M,
+    MODEL_360M,
     MODEL_410M,
     MODEL_450M,
+    MODEL_475M,
+    MODEL_558M,
+    MODEL_700M,
     MODEL_770M,
     MODEL_780M,
+    MODEL_950M,
     MODEL_0_3B,
     MODEL_0_5B,
+    MODEL_0_6B,
     MODEL_1B,
+    MODEL_1_2B,
     MODEL_1_3B,
     MODEL_1_4B,
+    MODEL_1_5B,
+    MODEL_1_6B,
+    MODEL_1_7B,
+    MODEL_1_8B,
     MODEL_2B,
+    MODEL_2_6B,
     MODEL_2_8B,
+    MODEL_2_9B,
     MODEL_3B,
     MODEL_4B,
     MODEL_6B,
@@ -54,17 +71,19 @@ enum e_model {
     MODEL_15B,
     MODEL_16B,
     MODEL_20B,
+    MODEL_27B,
     MODEL_30B,
     MODEL_32B,
     MODEL_34B,
     MODEL_35B,
+    MODEL_36B,
     MODEL_40B,
     MODEL_65B,
     MODEL_70B,
-    MODEL_106B_A12B,
+    MODEL_120B,
     MODEL_142B,
     MODEL_236B,
-    MODEL_355B_A32B,
+    MODEL_290B,
     MODEL_314B,
     MODEL_405B,
     MODEL_671B,
@@ -72,22 +91,33 @@ enum e_model {
     MODEL_MEDIUM,
     MODEL_LARGE,
     MODEL_XL,
+    MODEL_A1_7B,
     MODEL_A2_7B,
     MODEL_8x7B,
     MODEL_8x22B,
     MODEL_16x12B,
+    MODEL_16x3_8B,
     MODEL_10B_128x3_66B,
-    MODEL_21B_A3B, // Ernie MoE small
     MODEL_57B_A14B,
-    MODEL_27B,
     MODEL_17B_16E,
     MODEL_17B_128E,
-    MODEL_80B_A13B,
-    MODEL_300B_A47B, // Ernie MoE big
+    MODEL_A13B,
+    MODEL_7B_A1B,
+    MODEL_8B_A1B,
     MODEL_16B_A1B,
+    MODEL_21B_A3B, // Ernie MoE small
+    MODEL_30B_A3B,
+    MODEL_80B_A13B,
     MODEL_100B_A6B,
+    MODEL_106B_A12B,
+    MODEL_235B_A22B,
+    MODEL_300B_A47B, // Ernie MoE big
+    MODEL_355B_A32B,
+    MODEL_E2B,
+    MODEL_E4B,
 };
 
+
 struct llama_layer_nextn {
     struct ggml_tensor * eh_proj          = nullptr;
     struct ggml_tensor * embed_tokens     = nullptr;