Remove CUDA sidecar builds, keep CPU + Cloud only

CUDA sidecars are ~2GB and too slow to upload from the Windows runner.
Cloud (Deepgram) provides faster transcription anyway. Removed:

- CUDA build steps from Windows and Linux sidecar workflows
- CUDA option from the SidecarSetup download screen

Remaining sidecar variants:
- Cloud (Deepgram): ~50 MB - recommended for most users
- Local CPU: ~500 MB - for offline/privacy use

CUDA can be revisited once the managed Deepgram service is ready.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

src/lib/components/SidecarSetup.svelte

@@ -126,23 +126,6 @@
           </div>
         </label>
 
-        <label class="variant-option" class:selected={variant === "cuda"}>
-          <input
-            type="radio"
-            name="variant"
-            value="cuda"
-            bind:group={variant}
-          />
-          <div class="variant-info">
-            <span class="variant-name">Local - GPU (NVIDIA CUDA)</span>
-            <span class="variant-desc">~2 GB download</span>
-            <span class="variant-detail">
-              Runs Whisper AI models locally using your NVIDIA GPU for fast
-              transcription. No internet needed after download. Requires an
-              NVIDIA GPU with CUDA support.
-            </span>
-          </div>
-        </label>
       </div>
 
       <button class="download-btn" onclick={startDownload}>