engines: Test button inline with result + selectable; benchmark: persist WAV/ref paths (v2.0.7)

- Engines tab: move Test button out of the toolbar, place it in a row
  directly beside the result label so button and output are co-located;
  result label is now selectable so error text can be copied
- Benchmark tab: add bench_wav/bench_ref/bench_expand_models to Config;
  file pickers restore last-used paths on open; any change (file-set,
  toggle, or Run) writes directly to cfg so paths survive without Save;
  [benchmark] section written to config.toml on Save

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

linux/blitztext/__init__.py

@@ -6,4 +6,4 @@ counterpart to the macOS Blitztext menu bar app: it runs natively on the host
 (not in a container) so it can type into any application via xdotool.
 """
 
-__version__ = "2.0.6"
+__version__ = "2.0.7"

linux/blitztext/config.py

@@ -120,6 +120,10 @@ class Config:
     # Wakeword engines to compare in the benchmark (each a wyoming-openwakeword
     # endpoint). Empty → the benchmark falls back to the live [wakeword] above.
     wakeword_engines: list[WakewordEngine] = field(default_factory=list)
+    # STT benchmark: last-used WAV / reference transcript paths and options
+    bench_wav: str = ""
+    bench_ref: str = ""
+    bench_expand_models: bool = False
     # workflows
     workflows: list[Workflow] = field(default_factory=list)
 
@@ -236,6 +240,9 @@ def load(path: Path = CONFIG_PATH) -> Config:
         tts_api_key_env=tts.get("api_key_env", ""),
         tts_model=tts.get("model", ""),
         tts_voices=list(tts.get("voices", ["alloy", "echo", "fable", "onyx", "nova", "shimmer"])),
+        bench_wav=data.get("benchmark", {}).get("wav", ""),
+        bench_ref=data.get("benchmark", {}).get("ref", ""),
+        bench_expand_models=bool(data.get("benchmark", {}).get("expand_models", False)),
     )
 
     for entry in data.get("workflow", []):
@@ -371,6 +378,11 @@ def save(cfg: Config, path: Path = CONFIG_PATH) -> None:
             "model": cfg.tts_model,
             "voices": cfg.tts_voices,
         },
+        "benchmark": {
+            "wav": cfg.bench_wav,
+            "ref": cfg.bench_ref,
+            "expand_models": cfg.bench_expand_models,
+        },
         "wakeword_engine": [
             {"name": e.name, "uri": e.uri, "model": e.model} for e in cfg.wakeword_engines
         ],

linux/blitztext/gtksettings.py

@@ -1146,9 +1146,8 @@ notebook.bt-nb tab:checked label {
         qs.set_tooltip_text("Fill the form from a common service template")
         qs.connect("clicked", self._show_stt_templates)
         bar.pack_start(qs, False, False, 0)
-        # Right: actions
+        # Right: actions (Test moved to its own row below the config)
         for label, cb, tip in (
-            ("Test",    self._stt_test,                  "Record 4 s and transcribe to test this engine"),
             ("Delete",  self._stt_delete,                "Remove this engine preset"),
             ("⟳",       lambda _b: self._refresh_status(), "Re-check connection status"),
         ):
@@ -1185,8 +1184,19 @@ notebook.bt-nb tab:checked label {
                                     tooltip="Precision of the model. int8 is fastest and uses least memory. "
                                             "float16 is most accurate. Auto lets Blitztext decide.")
 
-        self.stt_result = Gtk.Label(xalign=0.0); self.stt_result.set_line_wrap(True)
-        box.pack_start(self.stt_result, False, False, 2)
+        test_row = Gtk.Box(spacing=10)
+        test_row.set_margin_top(6)
+        stt_test_btn = Gtk.Button(label="Test")
+        stt_test_btn.set_tooltip_text("Record 4 s and transcribe to test this engine")
+        stt_test_btn.connect("clicked", self._stt_test)
+        stt_test_btn.set_valign(Gtk.Align.START)
+        test_row.pack_start(stt_test_btn, False, False, 0)
+        self.stt_result = Gtk.Label(xalign=0.0)
+        self.stt_result.set_line_wrap(True)
+        self.stt_result.set_selectable(True)
+        self.stt_result.set_valign(Gtk.Align.START)
+        test_row.pack_start(self.stt_result, True, True, 0)
+        box.pack_start(test_row, False, False, 2)
         self._stt_load(self.stt_combo.get_active())
         return box
 
@@ -1842,17 +1852,31 @@ notebook.bt-nb tab:checked label {
         ft = Gtk.FileFilter(); ft.set_name("Text (.txt)"); ft.add_pattern("*.txt"); reff.add_filter(ft)
         self.bench_ref = _labeled(page, "Reference (.txt)", reff)
 
+        # Restore saved paths
+        if self.cfg.bench_wav and Path(self.cfg.bench_wav).exists():
+            wavf.set_filename(self.cfg.bench_wav)
+        if self.cfg.bench_ref and Path(self.cfg.bench_ref).exists():
+            reff.set_filename(self.cfg.bench_ref)
+
         def _on_wav_set(_b):
             fn = wavf.get_filename()
             if not fn: return
+            self.cfg.bench_wav = fn
             p = Path(fn)
             for ext in (".reference.txt", ".txt"):
                 cand = p.with_name(p.stem + ext)
                 if cand.exists():
                     reff.set_filename(str(cand))
+                    self.cfg.bench_ref = str(cand)
                     break
         wavf.connect("file-set", _on_wav_set)
 
+        def _on_ref_set(_b):
+            fn = reff.get_filename()
+            if fn:
+                self.cfg.bench_ref = fn
+        reff.connect("file-set", _on_ref_set)
+
         run_row = Gtk.Box(spacing=16)
         run_row.set_margin_bottom(2)
         run = Gtk.Button(label="Run benchmark"); run.connect("clicked", self._run_bench)
@@ -1862,6 +1886,9 @@ notebook.bt-nb tab:checked label {
         self.bench_expand.set_tooltip_text(
             "Fetch every available model from each remote engine and run a separate "
             "benchmark row per model — useful to compare models on the same server.")
+        self.bench_expand.set_active(self.cfg.bench_expand_models)
+        self.bench_expand.connect("toggled",
+            lambda cb: setattr(self.cfg, "bench_expand_models", cb.get_active()))
         run_row.pack_start(self.bench_expand, False, False, 0)
         page.pack_start(run_row, False, False, 6)
 
@@ -1945,6 +1972,11 @@ notebook.bt-nb tab:checked label {
     def _run_bench(self, _b) -> None:
         wav = self.bench_wav.get_filename()
         refp = self.bench_ref.get_filename()
+        # Persist immediately so paths survive without clicking Save
+        if wav:
+            self.cfg.bench_wav = wav
+        if refp:
+            self.cfg.bench_ref = refp
         if not wav or not refp:
             self._error("Pick a .wav file and a matching reference .txt file.")
             return