fix(desktop): resolve 2 release-blocking P1 defects

P1-04: GenerationPipeline hardcoded model="default" causing classroom
generation 404. Added model field to GenerationPipeline struct, passed
from kernel config via with_driver(driver, model). Static scene
generation now receives model parameter.

P1-03: LLM API concurrent 500 DATABASE_ERROR. Added transient DB error
retry (PoolTimedOut/Io) in create_relay_task with 200ms backoff.
Recommend setting ZCLAW_DB_MIN_CONNECTIONS=10 for burst resilience.

desktop/src-tauri/src/classroom_commands/generate.rs

@@ -132,7 +132,7 @@ pub async fn classroom_generate(
     let pipeline = {
         let ks = kernel_state.lock().await;
         if let Some(kernel) = ks.as_ref() {
-            GenerationPipeline::with_driver(kernel.driver())
+            GenerationPipeline::with_driver(kernel.driver(), kernel.config().model().to_string())
         } else {
             GenerationPipeline::new()
         }