Switch PLaMo translation from CUDA to Vulkan backend

[Copilot]

Migrates PLaMo translator plugin from CUDA 12 to Vulkan backend for broader GPU support.

Changes

• Package dependencies: LLamaSharp.Backend.Cuda12.Windows → LLamaSharp.Backend.Vulkan.Windows (0.25.0)
• Native library selection: NativeLibraryWithCuda(12, ...) → NativeLibraryWithVulkan(...)
• Build configuration: Removed CUDA DLL copy directives and CI toolkit installation step

// Before
yield return new NativeLibraryWithCuda(12, description.Library, description.AvxLevel, description.SkipCheck);

// After
yield return new NativeLibraryWithVulkan(description.Library, description.AvxLevel, description.SkipCheck);

CPU fallback via NativeLibraryWithAvx remains unchanged.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>PlaMo翻訳でCUDA版をやめてValkan版のみ対応にする</issue_title>
<issue_description></issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes PlaMo翻訳でCUDA版をやめてValkan版のみ対応にする #498

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.