Migrate amc4th training (#1316)

* Migrate amc4th training

* Refine RL example scripts

* Resolve PR comments

Co-authored-by: luocy16 <luocy16@mails.tsinghua.edu.cn>

examples/rl/experiment_config/backtest/twap.yml

@@ -0,0 +1,21 @@
+order_file: ./data/backtest_orders.csv
+start_time: "9:45"
+end_time: "14:44"
+qlib:
+  provider_uri_1min: ./data/bin
+  feature_root_dir: ./data/pickle
+  feature_columns_today: [
+    "$open", "$high", "$low", "$close", "$vwap", "$volume",
+  ]
+  feature_columns_yesterday: [
+    "$open_v1", "$high_v1", "$low_v1", "$close_v1", "$vwap_v1", "$volume_v1",
+  ]
+exchange:
+  limit_threshold: ['$close == 0', '$close == 0']
+  deal_price: ["If($close == 0, $vwap, $close)", "If($close == 0, $vwap, $close)"]
+  volume_threshold:
+    all: ["cum", "0.2 * DayCumsum($volume, '9:45', '14:44')"]
+    buy: ["current", "$close"]
+    sell: ["current", "$close"]
+strategies: {}  # Placeholder
+concurrency: 5