feat(reporting): share the report-tree writer between the CLI and the API

The per-section markdown report tree was written only by the CLI, so programmatic
(TradingAgentsGraph) runs produced no saved reports.

- Extract the writer into tradingagents/reporting.write_report_tree.
- The CLI's save_report_to_disk delegates to it (no behavior change).
- Add TradingAgentsGraph.save_reports(final_state, ticker) so headless/API callers
  get the same report tree, defaulting under results_dir.

tradingagents/graph/trading_graph.py

@@ -32,6 +32,7 @@ from tradingagents.dataflows.config import set_config
 from tradingagents.dataflows.utils import safe_ticker_component
 from tradingagents.default_config import DEFAULT_CONFIG
 from tradingagents.llm_clients import create_llm_client
+from tradingagents.reporting import write_report_tree
 
 from .checkpointer import checkpoint_step, clear_checkpoint, get_checkpointer, thread_id
 from .conditional_logic import ConditionalLogic
@@ -358,6 +359,21 @@ class TradingAgentsGraph:
                 self._checkpointer_ctx = None
                 self.graph = self.workflow.compile()
 
+    def save_reports(self, final_state, ticker, save_path=None) -> Path:
+        """Write the markdown report tree for a completed run, like the CLI does.
+
+        Programmatic callers get the same on-disk reports the CLI produces. Pass
+        an explicit ``save_path`` or let it default under ``results_dir``.
+        """
+        if save_path is None:
+            stamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+            save_path = (
+                Path(self.config["results_dir"])
+                / "reports"
+                / f"{safe_ticker_component(ticker)}_{stamp}"
+            )
+        return write_report_tree(final_state, ticker, save_path)
+
     def _run_graph(self, company_name, trade_date, asset_type: str = "stock"):
         """Execute the graph and write the resulting state to disk and memory log."""
         # Initialize state — inject memory log context for PM and the