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/config/agent.rs

@@ -531,6 +531,14 @@ impl Agent {
         self.config.compression_threshold
     }
 
+    pub fn max_tool_result_chars(&self) -> Option<usize> {
+        self.config.max_tool_result_chars
+    }
+
+    pub fn compression_keep_last(&self) -> Option<usize> {
+        self.config.compression_keep_last
+    }
+
     pub fn is_dynamic_instructions(&self) -> bool {
         self.config.dynamic_instructions
     }
@@ -679,6 +687,10 @@ pub struct AgentConfig {
     pub memory: Option<bool>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub compression_threshold: Option<usize>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub max_tool_result_chars: Option<usize>,
+    #[serde(default, skip_serializing_if = "Option::is_none")]
+    pub compression_keep_last: Option<usize>,
     #[serde(default)]
     pub description: String,
     #[serde(default)]