refactor(kon): remove LLM command stubs — will be rewritten with llama-cpp-rs in Phase 3

- Remove 8 LLM command registrations from invoke_handler macro
- Delete commands/llm.rs stub file
- Remove pub mod llm from commands/mod.rs
- Build passes cleanly with cargo clippy

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

src-tauri/src/commands/llm.rs

@@ -1,53 +0,0 @@
-// LLM stubs — ONNX-based local LLM coming post-revenue.
-
-#[tauri::command]
-pub fn check_llm_engine() -> Result<bool, String> {
-    Ok(false)
-}
-
-#[tauri::command]
-pub fn check_llm_model(_size: String) -> Result<bool, String> {
-    Ok(false)
-}
-
-#[tauri::command]
-pub fn list_llm_models() -> Result<Vec<String>, String> {
-    Ok(vec![])
-}
-
-#[tauri::command]
-pub async fn download_llm_model(
-    _app: tauri::AppHandle,
-    _size: String,
-) -> Result<String, String> {
-    Err("LLM not available yet — coming soon".into())
-}
-
-#[tauri::command]
-pub async fn load_llm_model(
-    _size: String,
-) -> Result<String, String> {
-    Err("LLM not available yet — coming soon".into())
-}
-
-#[tauri::command]
-pub async fn llm_generate(
-    _prompt: String,
-    _max_tokens: Option<u32>,
-) -> Result<String, String> {
-    Err("LLM not available yet — coming soon".into())
-}
-
-#[tauri::command]
-pub async fn llm_extract_tasks(
-    _transcript: String,
-) -> Result<Vec<String>, String> {
-    Err("LLM not available yet — coming soon".into())
-}
-
-#[tauri::command]
-pub async fn llm_format_transcript(
-    _transcript: String,
-) -> Result<String, String> {
-    Err("LLM not available yet — coming soon".into())
-}

src-tauri/src/commands/mod.rs

@@ -1,7 +1,6 @@
 pub mod audio;
 pub mod clipboard;
 pub mod hardware;
-pub mod llm;
 pub mod models;
 pub mod transcription;
 pub mod windows;

src-tauri/src/lib.rs

@@ -75,15 +75,6 @@ pub fn run() {
             // Hardware
             commands::hardware::probe_system,
             commands::hardware::rank_models,
-            // LLM stubs
-            commands::llm::download_llm_model,
-            commands::llm::check_llm_model,
-            commands::llm::list_llm_models,
-            commands::llm::load_llm_model,
-            commands::llm::check_llm_engine,
-            commands::llm::llm_generate,
-            commands::llm::llm_extract_tasks,
-            commands::llm::llm_format_transcript,
         ])
         .run(tauri::generate_context!())
         .expect("error while running Kon");