Fix CUDA extension compilation and CPU fallback logic

[Electroiscoding]

This PR fixes several issues related to the CUDA extension:

1. Relaxed the CUDA_HOME requirement in setup.py to allow the CUDA extension to compile as long as nvcc is available, addressing environments where PyTorch or CUDA_HOME is partially set up.
2. Removed an unused variable in the tokenize_kernel inside gpu_engine_cuda.cu to prevent compilation warnings/errors.
3. Added robust error handling in vocabulary.py to appropriately trigger the CPU fallback if the CUDA backend initialized but found no hardware, preventing runtime errors later.
4. Corrected the is_cuda_available() method in c_ext/__init__.py to accurately reflect a lack of CUDA hardware, setting _cuda_error instead of silently claiming availability.

---

PR created automatically by Jules for task 2234727710909910041 started by @Electroiscoding

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.