New models (#15861)

* mlx: add laguna model support

* convert: support fp8 safetensors import

Decode HF F8_E4M3 safetensors with block scale companions into GGUF-supported tensor types, and record which output tensors came from FP8 source weights.

Use that source-precision metadata during create quantization: default FP8-sourced GGUFs to Q8_0, keep non-FP8 tensors at their original precision for Q8_0, and promote non-FP8 quantizable tensors to Q8_0 for Q4_K requests.

* ggml: add laguna model support

* server: preserve generate logprobs with builtin parsers

Generate requests were dropping logprob-only chunks whenever a builtin parser buffered visible content. Chat already handled this case, but generate only forwarded chunks with visible response, thinking, or tool-call output.

Keep generate chunks that carry logprobs even when the builtin parser has not flushed visible content yet, and add a regression test that exercises the behavior with a generic thinking parser.

* review comments - perf improvements

* ggml: implement nemotron 3 nano omni

* add poolside integration

* update poolside doc

* adapt to new cache setup

* fix test

* fix test

---------

Co-authored-by: Eva Ho <hoyyeva@gmail.com>

fs/ggml/ggml.go

@@ -283,10 +283,11 @@ func (kv KV) OllamaEngineRequired() bool {
 		"gemma3n",
 		"gemma4",
 		"gptoss", "gpt-oss",
+		"laguna",
 		"llama4",
 		"mistral3",
 		"mllama",
-		"nemotron_h", "nemotron_h_moe",
+		"nemotron_h", "nemotron_h_moe", "nemotron_h_omni",
 		"nomic-bert",
 		"olmo3",
 		"qwen25vl",
@@ -897,7 +898,7 @@ func (f GGML) FlashAttention() bool {
 		"lfm2",
 		"lfm2moe",
 		"mistral3",
-		"nemotron_h", "nemotron_h_moe",
+		"nemotron_h", "nemotron_h_moe", "nemotron_h_omni",
 		"olmo3",
 		"qwen3", "qwen3moe",
 		"qwen35", "qwen35moe",