Update LLM support for llama.cpp and enhance configuration management

- Added support for `llamacpp-models.yaml` in `.gitignore` and implemented logic to copy it in `gpu-rent.ps1` and `gpu-rent.sh`.
- Enhanced CLI to prompt for llama.cpp model presets during setup and execution, improving user experience.
- Updated configuration handling to include `llamacpp_models_manifest` and related functions for managing llama.cpp models.
- Improved documentation in `cli.md` and `llm.md` to reflect changes in llama.cpp integration and model management.
- Refactored provisioning logic to handle llama.cpp model downloads and configurations effectively.

src/gpu_rent/cli.py

@@ -436,6 +436,7 @@ def up(
         except ValueError as exc:
             raise GpuRentError(str(exc)) from exc
 
+        asked_model_preset = False
         if not yes and runtime == "none" and not llm and not ollama and not llamacpp:
             choice = typer.prompt(
                 "Поднять LLM рядом со SwarmUI? [none/ollama/llamacpp]",
@@ -457,6 +458,53 @@ def up(
                     write_ollama_models_preset(
                         cfg.ollama_models_manifest, preset.strip().lower()
                     )
+                asked_model_preset = True
+            elif runtime == "llamacpp":
+                from gpu_rent.llm_runtime import (
+                    LLAMACPP_PRESET_HELP,
+                    ensure_llamacpp_manifest_from_example,
+                    write_llamacpp_models_preset,
+                )
+
+                ensure_llamacpp_manifest_from_example()
+                console.print(LLAMACPP_PRESET_HELP)
+                preset = typer.prompt(
+                    "llama.cpp GGUF preset [recommended/light/stock/empty]",
+                    default="recommended",
+                )
+                if preset.strip().lower() not in {"keep", "example"}:
+                    write_llamacpp_models_preset(
+                        cfg.llamacpp_models_manifest, preset.strip().lower()
+                    )
+                asked_model_preset = True
+
+        # Runtime уже в vars (напр. llamacpp) — всё равно спросить модель, default=keep.
+        if not yes and not asked_model_preset and runtime == "llamacpp":
+            from gpu_rent.llm_runtime import (
+                LLAMACPP_PRESET_HELP,
+                ensure_llamacpp_manifest_from_example,
+                write_llamacpp_models_preset,
+            )
+
+            ensure_llamacpp_manifest_from_example()
+            console.print(LLAMACPP_PRESET_HELP)
+            preset = typer.prompt(
+                "llama.cpp GGUF preset [recommended/light/stock/empty/keep]",
+                default="keep",
+            )
+            key = preset.strip().lower()
+            if key not in {"keep", "example", ""}:
+                write_llamacpp_models_preset(cfg.llamacpp_models_manifest, key)
+        elif not yes and not asked_model_preset and runtime == "ollama":
+            ensure_ollama_manifest_from_example()
+            console.print(PRESET_HELP)
+            preset = typer.prompt(
+                "Ollama preset [recommended/light/stock/alt/empty/keep]",
+                default="keep",
+            )
+            key = preset.strip().lower()
+            if key not in {"keep", "example", ""}:
+                write_ollama_models_preset(cfg.ollama_models_manifest, key)
 
         cfg = replace(cfg, llm_runtime=runtime)
         if runtime != "none":
@@ -465,6 +513,9 @@ def up(
         def confirm(msg: str) -> bool:
             return typer.confirm(msg)
 
+        def ask(msg: str, default: str = "") -> str:
+            return typer.prompt(msg, default=default)
+
         state = cmd_up(
             cfg,
             no_spot=no_spot,
@@ -473,6 +524,7 @@ def up(
             adopt=adopt,
             update=False if no_update else None,
             confirm=confirm,
+            ask=None if yes else ask,
             log=lambda m: console.print(m),
         )
         if no_tunnel: