fix(routines): resolve message tool channel/target from per-job metadata (#708)

* fix(routines): resolve message tool channel/target from per-job metadata

When a routine's notify.channel is None, the message tool had no way to
resolve channel/target for full-job workers, causing "No target specified"
errors. The previous approach mutated shared global state via
set_message_tool_context(), which also raced with concurrent jobs.

Now the routine's notify config (channel + user) is carried in the job's
metadata JSON, and MessageTool::execute falls back to ctx.metadata when
neither explicit params nor conversation defaults are available. This
eliminates both the None-channel bug and the concurrent-job race.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* style: apply cargo fmt

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(message): broadcast to all channels when notify.channel is None

Address review feedback:
- Fix stale "see above" comment → "populated below"
- When notify.channel is None, use broadcast_all instead of erroring
  with "No channel specified". This matches NotifyConfig semantics
  where channel=None means "broadcast to all channels"
- Channel resolution is now Option<String>: param → default → metadata → None
- When None, MessageTool uses ChannelManager::broadcast_all(target, response)
  and reports which channels succeeded/failed
- Add regression test for broadcast-all behavior

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: use failed channels in error message, remove redundant comment

Address review feedback:
- Use `failed` vec in error message instead of re-querying channel_names
- Remove redundant orphaned comment block in routine_engine.rs

Co-Authored-By: Claude Opus 4.6 <[email protected]>

---------

Co-authored-by: Claude Opus 4.6 <[email protected]>

src/agent/routine_engine.rs

@@ -492,18 +492,13 @@ async fn execute_full_job(
             reason: "scheduler not available".to_string(),
         })?;
 
-    // Set the message tool's default channel/target from the routine's notify config
-    // so the LLM can send results without triggering cross-channel approval.
-    // TODO: This mutates shared global state and can race with concurrent jobs.
-    // Move notify config into JobContext metadata and apply per-job instead.
+    let mut metadata = serde_json::json!({ "max_iterations": max_iterations });
+    // Carry the routine's notify config in job metadata so the message tool
+    // can resolve channel/target per-job without global state mutation.
     if let Some(channel) = &routine.notify.channel {
-        scheduler
-            .tools()
-            .set_message_tool_context(Some(channel.clone()), Some(routine.notify.user.clone()))
-            .await;
+        metadata["notify_channel"] = serde_json::json!(channel);
     }
-
-    let metadata = serde_json::json!({ "max_iterations": max_iterations });
+    metadata["notify_user"] = serde_json::json!(&routine.notify.user);
 
     // Build approval context: UnlessAutoApproved tools are auto-approved for routines;
     // Always tools require explicit listing in tool_permissions.