refactor: rewrite to CoinHunter V2 flat architecture

Replace the V1 commands/services split with a flat, direct architecture:
- cli.py dispatches directly to service functions
- New services: account, market, trade, opportunity
- Thin Binance wrappers: spot_client, um_futures_client
- Add audit logging, runtime paths, and TOML config
- Remove legacy V1 code: commands/, precheck, review engine, smart executor
- Add ruff + mypy toolchain and fix edge cases in trade params

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

tests/test_cli.py

@@ -1,58 +1,38 @@
-"""Tests for CLI routing and parser behavior."""
+"""CLI tests for CoinHunter V2."""
 
-import pytest
+from __future__ import annotations
 
-from coinhunter.cli import ALIASES, MODULE_MAP, build_parser, run_python_module
+import io
+import unittest
+from unittest.mock import patch
+
+from coinhunter import cli
 
 
-class TestAliases:
-    def test_all_aliases_resolve_to_canonical(self):
-        for alias, canonical in ALIASES.items():
-            assert canonical in MODULE_MAP, f"alias {alias!r} points to missing canonical {canonical!r}"
-
-    def test_no_alias_is_itself_an_alias_loop(self):
-        for alias in ALIASES:
-            assert alias not in ALIASES.values() or alias in MODULE_MAP
-
-
-class TestModuleMap:
-    def test_all_modules_exist(self):
-        import importlib
-
-        for command, module_name in MODULE_MAP.items():
-            full = f"coinhunter.{module_name}"
-            mod = importlib.import_module(full)
-            assert hasattr(mod, "main"), f"{full} missing main()"
-
-
-class TestBuildParser:
-    def test_help_includes_commands(self):
-        parser = build_parser()
+class CLITestCase(unittest.TestCase):
+    def test_help_includes_v2_commands(self):
+        parser = cli.build_parser()
         help_text = parser.format_help()
-        assert "coinhunter diag" in help_text
-        assert "coinhunter exec" in help_text
+        self.assertIn("init", help_text)
+        self.assertIn("account", help_text)
+        self.assertIn("opportunity", help_text)
 
-    def test_parses_command_and_args(self):
-        parser = build_parser()
-        ns = parser.parse_args(["exec", "bal"])
-        assert ns.command == "exec"
-        assert "bal" in ns.args
+    def test_init_dispatches(self):
+        captured = {}
+        with patch.object(cli, "ensure_init_files", return_value={"force": True, "root": "/tmp/ch"}), patch.object(
+            cli, "print_json", side_effect=lambda payload: captured.setdefault("payload", payload)
+        ):
+            result = cli.main(["init", "--force"])
+        self.assertEqual(result, 0)
+        self.assertTrue(captured["payload"]["force"])
 
-    def test_version_action_exits(self):
-        parser = build_parser()
-        with pytest.raises(SystemExit) as exc:
-            parser.parse_args(["--version"])
-        assert exc.value.code == 0
+    def test_old_command_is_rejected(self):
+        with self.assertRaises(SystemExit):
+            cli.main(["exec", "bal"])
 
-
-class TestRunPythonModule:
-    def test_runs_module_main_and_returns_int(self):
-        result = run_python_module("commands.paths", [], "coinhunter paths")
-        assert result == 0
-
-    def test_mutates_sys_argv(self):
-        import sys
-
-        original = sys.argv[:]
-        run_python_module("commands.paths", ["--help"], "coinhunter paths")
-        assert sys.argv == original
+    def test_runtime_error_is_rendered_cleanly(self):
+        stderr = io.StringIO()
+        with patch.object(cli, "load_config", side_effect=RuntimeError("boom")), patch("sys.stderr", stderr):
+            result = cli.main(["market", "tickers", "BTCUSDT"])
+        self.assertEqual(result, 1)
+        self.assertIn("error: boom", stderr.getvalue())