Add persian-tutor: Gradio-based GCSE Persian language learning app

Vocabulary study with FSRS spaced repetition, AI tutoring (Ollama/Claude),
essay marking, idioms browser, Anki export, and dashboard. 918 vocabulary
entries across 39 categories. 41 tests passing.

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

python/persian-tutor/tests/test_ai.py

@@ -0,0 +1,89 @@
+"""Tests for ai.py — dual AI backend."""
+
+import sys
+from pathlib import Path
+from unittest.mock import patch, MagicMock
+
+import pytest
+
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+import ai
+
+
+def test_ask_ollama_calls_ollama_chat():
+    """ask_ollama should call ollama.chat with correct messages."""
+    mock_response = MagicMock()
+    mock_response.message.content = "test response"
+
+    with patch("ai.ollama.chat", return_value=mock_response) as mock_chat:
+        result = ai.ask_ollama("Hello", system="Be helpful")
+        assert result == "test response"
+
+        call_args = mock_chat.call_args
+        messages = call_args.kwargs.get("messages") or call_args[1].get("messages")
+        assert len(messages) == 2
+        assert messages[0]["role"] == "system"
+        assert messages[1]["role"] == "user"
+        assert messages[1]["content"] == "Hello"
+
+
+def test_ask_ollama_no_system():
+    """ask_ollama without system prompt should only send user message."""
+    mock_response = MagicMock()
+    mock_response.message.content = "response"
+
+    with patch("ai.ollama.chat", return_value=mock_response) as mock_chat:
+        ai.ask_ollama("Hi")
+        call_args = mock_chat.call_args
+        messages = call_args.kwargs.get("messages") or call_args[1].get("messages")
+        assert len(messages) == 1
+        assert messages[0]["role"] == "user"
+
+
+def test_ask_claude_calls_subprocess():
+    """ask_claude should call claude CLI via subprocess."""
+    with patch("ai.subprocess.run") as mock_run:
+        mock_run.return_value = MagicMock(stdout="Claude says hi\n")
+        result = ai.ask_claude("Hello")
+        assert result == "Claude says hi"
+        mock_run.assert_called_once()
+        args = mock_run.call_args[0][0]
+        assert args[0] == "claude"
+        assert "-p" in args
+
+
+def test_ask_fast_uses_ollama():
+    """ask with quality='fast' should use Ollama."""
+    with patch("ai.ask_ollama", return_value="ollama response") as mock:
+        result = ai.ask("test", quality="fast")
+        assert result == "ollama response"
+        mock.assert_called_once()
+
+
+def test_ask_smart_uses_claude():
+    """ask with quality='smart' should use Claude."""
+    with patch("ai.ask_claude", return_value="claude response") as mock:
+        result = ai.ask("test", quality="smart")
+        assert result == "claude response"
+        mock.assert_called_once()
+
+
+def test_chat_ollama():
+    """chat_ollama should pass multi-turn messages."""
+    mock_response = MagicMock()
+    mock_response.message.content = "continuation"
+
+    with patch("ai.ollama.chat", return_value=mock_response) as mock_chat:
+        messages = [
+            {"role": "user", "content": "Hi"},
+            {"role": "assistant", "content": "Hello!"},
+            {"role": "user", "content": "How are you?"},
+        ]
+        result = ai.chat_ollama(messages, system="Be helpful")
+        assert result == "continuation"
+
+        call_args = mock_chat.call_args
+        all_msgs = call_args.kwargs.get("messages") or call_args[1].get("messages")
+        # system + 3 conversation messages
+        assert len(all_msgs) == 4