feat: OpenVINO CPU MetricGAN mask; 100 ms hop

Per-worker OpenVINO CPU runs the MetricGAN BLSTM mask; SpeechBrain
keeps STFT/ISTFT. ENHANCE_INFER=auto with torch fallback. Hop is
0.1 s after isolated last_ms ~6 ms (model+ov+beamform).

tests/test_config.py

@@ -13,3 +13,7 @@ def test_default_beamform_mode_auto():
 
 def test_torch_num_threads_default_one():
     assert Config().torch_num_threads == 1
+
+
+def test_enhance_infer_default_auto():
+    assert Config().enhance_infer == "auto"