feat: long-running session improvements

- Hang fix: 120s timeout on compression LLM call + raw_stream
  channel-close handling (None => break instead of spinning)
- Supervisor effective active count: stop counting finished
  JoinHandles as occupying capacity slots
- Universal tool result size cap: truncate_if_needed() on
  ToolResult, applied in eval_tool_calls() after escalation block;
  configurable via max_tool_result_chars in AppConfig/AgentConfig
- Windowed compression: compression_keep_last config param keeps
  the N most recent messages visible after compression
- Fix pre-existing flaky test: add #[serial] to
  handle_list_available_unrestricted_when_no_whitelist so it does
  not race with TestConfigDirGuard-based tests that temporarily
  populate the agents data dir

src/function/mod.rs

@@ -184,6 +184,20 @@ pub async fn eval_tool_calls(
         }
     }
 
+    {
+        let max_chars = ctx
+            .agent
+            .as_ref()
+            .and_then(|a| a.max_tool_result_chars())
+            .or_else(|| ctx.app.config.max_tool_result_chars);
+        if let Some(max_chars) = max_chars.filter(|&n| n > 0) {
+            output = output
+                .into_iter()
+                .map(|r| r.truncate_if_needed(max_chars))
+                .collect();
+        }
+    }
+
     Ok(output)
 }
 
@@ -219,6 +233,17 @@ impl ToolResult {
             thinking: vec![],
         }
     }
+
+    pub fn truncate_if_needed(mut self, max_chars: usize) -> Self {
+        let s = self.output.to_string();
+        if s.len() > max_chars {
+            let prefix = s.get(..max_chars).unwrap_or(s.as_str());
+            self.output = json!(format!(
+                "[truncated: tool output exceeded {max_chars} chars]\n{prefix}"
+            ));
+        }
+        self
+    }
 }
 
 #[derive(Debug, Clone, Default)]