feat: rede_korrigieren.py -m/--model zur Modellwahl

Das Modell steckt in den Cache-Schlüsseln; ein anderes Modell rechnet neu.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tests/test_rede_korrigieren.py

@@ -46,6 +46,8 @@ def no_model_calls(*args, **kwargs):
 
 def run_main(monkeypatch, *args):
     monkeypatch.setattr(sys, "argv", ["rede_korrigieren.py", *map(str, args)])
+    # main() sets the module-wide model; restore it after the test
+    monkeypatch.setattr(rk, "MODEL", rk.MODEL)
     rk.main()
 
 
@@ -301,6 +303,40 @@ def test_main_vtt_input_matches_txt_input(sample_copy, monkeypatch):
     ) == expected.read_text(encoding="utf-8")
 
 
+def test_main_model_option(sample_copy, monkeypatch):
+    default = rk.MODEL
+    used = []
+
+    def correct(chunk, index, system):
+        used.append(rk.MODEL)
+        return chunk.strip()
+
+    monkeypatch.setattr(rk, "correct_chunk", correct)
+    monkeypatch.setattr(rk, "find_paragraph_starts", every_nth_start(5))
+    parts = sample_copy.parent / "ramstein.de.korrigiert.parts"
+
+    run_main(monkeypatch, "-i", sample_copy)
+    default_parts = {p.name for p in parts.glob("part*")}
+    run_main(monkeypatch, "-i", sample_copy, "-m", "anderes:7b")
+
+    assert set(used) == {default, "anderes:7b"}
+    # Another model must not reuse the cached parts of the default model
+    assert {p.name for p in parts.glob("part*")} - default_parts
+
+
+def test_correct_chunk_sends_model(monkeypatch):
+    sent = {}
+
+    def urlopen(request, timeout=None):
+        sent.update(json.loads(request.data))
+        return FakeResponse(stream_body(["Text."]))
+
+    monkeypatch.setattr(rk, "MODEL", "anderes:7b")
+    monkeypatch.setattr(rk.urllib.request, "urlopen", urlopen)
+    rk.correct_chunk("Text.", 1, "system")
+    assert sent["model"] == "anderes:7b"
+
+
 def test_main_warns_on_word_count_deviation(sample_copy, monkeypatch, capsys):
     def shortening(chunk, index, system):
         words = chunk.split()