Fix RAG vector-space mismatch and search_rag retrieval quality

Queries embedded via llama-swap were searching corpora embedded via
Ollama (measured: spaces diverged). Introduce LocalLlm — the local
Ollama + llama-swap pair with LLM_BACKEND dispatch baked in — and route
all embedding writers through it; anything embedding via a concrete
client reintroduces the bug.

- search_rag: embed the model's query verbatim (no metadata boilerplate),
  make date optional — no time-decay when omitted, so "when did X
  happen?" queries rank purely by similarity across all time
- reembed_embeddings bin: re-embed summaries / calendar / search /
  knowledge entities via the active backend, with old-new cosine report
  per table and truncate-and-retry for inputs over the embed server's
  physical batch size
- import_calendar, import_search_history: embed through LocalLlm
- search_messages / get_sms_messages: render sender → recipient so sent
  messages are attributable to a conversation
- insight job failures: store the one-line anyhow context chain ({:#})
  instead of the Debug dump the client was shown verbatim
- serialize env_dispatch tests behind a lock (parallel-runner flake)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/ai/handlers.rs

@@ -510,7 +510,9 @@ pub async fn generate_insight_handler(
             }
             Ok(Ok(Err(e))) => {
                 log::error!("Insight generation failed for {}: {:?}", path, e);
-                if let Err(err) = dao.fail_job(&ctx, job_id, &format!("{:?}", e)) {
+                // `{:#}` = one-line context chain; the job's error_message is
+                // returned to the client verbatim, so no Debug/backtrace here.
+                if let Err(err) = dao.fail_job(&ctx, job_id, &format!("{:#}", e)) {
                     log::error!("Failed to mark job {} as failed: {:?}", job_id, err);
                 }
             }
@@ -884,7 +886,9 @@ pub async fn generate_agentic_insight_handler(
             }
             Ok(Ok(Err(e))) => {
                 log::error!("Agentic insight generation failed for {}: {:?}", path, e);
-                if let Err(err) = dao.fail_job(&ctx, job_id, &format!("{:?}", e)) {
+                // `{:#}` = one-line context chain; the job's error_message is
+                // returned to the client verbatim, so no Debug/backtrace here.
+                if let Err(err) = dao.fail_job(&ctx, job_id, &format!("{:#}", e)) {
                     log::error!("Failed to mark job {} as failed: {:?}", job_id, err);
                 }
             }