feat(#6): startup session picker for bare TUI mode

v1 coverage-audit slice b2. The audit found list_sessions had no
caller — the startup session picker (design-brief §4) was never built;
bare TUI mode was a hard usage error. Add SessionPickerApp (mirrors
AgentPickerApp) and resolve bare mode in _resolve_then_run.

- Bare TUI mode (no --session/--new) now valid → session picker.
  Resolution: 0 sessions -> [no_sessions] exit 14 (resume-only per
  §4 "no in-app creation, --new only"); exactly 1 -> auto-resume
  (§4 "picker only when >1"); >=2 -> SessionPickerApp -> resume pick
  (Esc/Ctrl-D -> exit 0).
- cli._parse: bare TUI valid; --send still requires one flag; --agent
  forbidden in bare mode. run_tui PRE-002 xor -> mutually-exclusive.
- Contract #6 amended (SessionPickerApp + bare-mode resolution) +
  validated. TDD: 3 picker pilot tests + 5 resolution tests + 3 cli
  validation tests. Suite 528 green; touched code ruff-clean.

Design note: bare + 0 sessions errors (honors §4's no-in-app-creation
clause); the friendlier auto-fall-through-to-new is deferred pending
operator preference.

tests/test_cli.py

@@ -175,10 +175,29 @@ class TestParseArgs:
             )
 
     def test_usage_neither_session_nor_new(self) -> None:
-        """usage_neither_session_nor_new: neither flag → UsageError('pass exactly one')."""
-        with pytest.raises(UsageError, match="pass exactly one"):
+        """usage_neither_session_nor_new: --send with neither flag → UsageError.
+
+        --send is non-interactive (no picker can open), so a session must be
+        named. Bare TUI mode (no --send) is now valid → session picker (§4).
+        """
+        with pytest.raises(UsageError, match="--send requires"):
             _parse_args(["--send", "hi", "--api-key", "k"])
 
+    def test_bare_tui_mode_accepted(self) -> None:
+        """bare_tui_mode (slice b2): no --send, no --session, no --new → valid;
+        _resolve_then_run drives the startup session picker (design-brief §4)."""
+        args = _parse_args(["--api-key", "k"])
+        assert args.send_content is None
+        assert args.session_id is None
+        assert args.new is False
+        assert args.agent_id is None
+
+    def test_usage_bare_tui_with_agent(self) -> None:
+        """bare_tui_with_agent (slice b2): bare TUI + --agent → UsageError
+        (--agent belongs with --new; bare mode opens the resume picker)."""
+        with pytest.raises(UsageError, match="belongs with --new"):
+            _parse_args(["--agent", "mimir", "--api-key", "k"])
+
     def test_usage_send_new_without_agent(self) -> None:
         """send_new_without_agent (issue #8): --send --new without --agent → UsageError.
 
@@ -1319,10 +1338,15 @@ class TestMain:
         rc = main(["--send", "hi", "--new", "--agent", "m", "--api-key", "k"])
         assert rc == 0
 
-    def test_usage_error_no_send(
+    def test_empty_argv_fails_on_auth(
         self, monkeypatch: pytest.MonkeyPatch, capsys: pytest.CaptureFixture[str]
     ) -> None:
-        """usage_error_no_send: empty argv → exit 10; stderr [usage_error]; _amain never called."""
+        """empty argv → exit 11 [auth_error]; _amain never called.
+
+        Since slice b2 bare TUI mode (no --send/--session/--new) is VALID (it
+        opens the session picker), so empty argv is no longer a usage error —
+        it now fails on the missing API key instead (still before _amain).
+        """
         amain_calls: list[int] = []
 
         async def fake_amain(args: ParsedArgs) -> int:
@@ -1331,8 +1355,8 @@ class TestMain:
 
         monkeypatch.setattr(cli_mod, "_amain", fake_amain)
         rc = main([])
-        assert rc == 10
-        assert "[usage_error]" in capsys.readouterr().err
+        assert rc == 11
+        assert "[auth_error]" in capsys.readouterr().err
         assert amain_calls == []
 
     def test_usage_error_both_session_and_new(