Fix clap detection silently analyzing the wrong audio track

Root cause of the reported 'timestamps off by different amounts,
sometimes early, sometimes late' bug: _extract_audio_wav() never
passed an explicit -map to ffmpeg, so on multi-audio-track files it
relied on ffmpeg's own default-stream heuristic -- which has no
knowledge of, and isn't guaranteed to agree with, whichever audio
track mpv is actually configured to play. A wrong-track clap is
essentially unrelated to the true one (not a small timestamp-math
error), which explains the inconsistent, file-dependent symptom --
and our single-audio-track synthetic fixtures never exercised this
path at all, since they only ever had one track to pick.

Fixed by threading an explicit ffmpeg stream index through
detect_claps() -> _extract_audio_wav(), sourced in main.py from mpv's
own track-list (the ff-index of whichever audio track has
selected=True). Detection now always analyzes the exact track the
user is listening to, on both 'Detect All' and 'Detect Next'.

Also: _probe_audio_start_time() used to silently fall back to 0.0 if
ffprobe couldn't parse start_time as a plain float (e.g. N/A) -- same
failure class (silently wrong instead of loud). Now emits a Python
warning instead, so a genuinely unparseable/unusual container is at
least visible rather than silently mis-corrected.

Verified two ways: (1) ruled out an initial alternate hypothesis
(video stream's own start_time mattering) empirically against the
existing single-track fixtures before ruling it out. (2) Built a new
two-audio-track fixture (sync_test_MULTITRACK.mp4: track 1 has a beep
at 3s, track 2 at 7s) and confirmed end-to-end through the real app --
switching mpv's active track from aid=1 to aid=2 correctly flips which
clap gets detected, which is exactly the mechanism this bug required.

scripts/smoke_test_multitrack_clap_detection.py

@@ -0,0 +1,88 @@
+"""Verifies clap detection analyzes whichever audio track mpv is actually
+playing, not whichever one ffmpeg's own default-stream heuristic would
+pick. Uses a synthetic fixture with two audio tracks with claps at very
+different times (3s vs 7s) -- if detection ignored mpv's active track
+selection, switching mpv to the second track would still incorrectly
+report the first track's 3s clap instead of the second track's 7s one.
+
+This directly tests the bug hypothesis from user report: 'sometimes too
+early, sometimes too late, by different amounts' -- exactly what you'd
+see if detection sometimes analyzed the wrong audio track on
+multi-audio-track files, since a wrong-track clap position is essentially
+unrelated to the true one (not a small timestamp-math error).
+"""
+import sys
+import time
+from pathlib import Path
+
+from PyQt6.QtWidgets import QApplication
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+import main as app_main
+
+ROOT = Path(__file__).resolve().parent.parent
+CLIP = str(ROOT / "test_media" / "sync_test_MULTITRACK.mp4")
+
+
+def wait_for(predicate, qapp, timeout=20):
+    deadline = time.time() + timeout
+    while not predicate() and time.time() < deadline:
+        qapp.processEvents()
+        time.sleep(0.05)
+    return predicate()
+
+
+def main():
+    qapp = QApplication(sys.argv)
+    win = app_main.MainWindow()
+    win.show()
+    qapp.processEvents()
+
+    win._load(CLIP)
+    qapp.processEvents()
+    time.sleep(0.3)
+    qapp.processEvents()
+
+    mpv = win.source_panel.player.mpv
+    print(f"default aid: {mpv.aid}")
+    default_index = win._current_audio_stream_index()
+    print(f"resolved ff-index for default track: {default_index}")
+    assert default_index == 1, f"expected default track's ff-index to be 1, got {default_index}"
+
+    win.detect_all_button.click()
+    qapp.processEvents()
+    ok = wait_for(lambda: win.detect_all_button.isEnabled(), qapp)
+    assert ok, "detection did not finish in time"
+    markers = win.source_panel.marker_strip.markers()
+    print(f"markers with default track selected: {markers}")
+    assert any(abs(m - 3.0) < 0.5 for m in markers), f"expected a marker near 3.0s (default track's clap): {markers}"
+    assert not any(abs(m - 7.0) < 0.5 for m in markers), f"should not see the OTHER track's 7.0s clap yet: {markers}"
+
+    # Now switch mpv to the second audio track (aid=2), simulating a file
+    # where the user is listening to a non-default track, and rerun
+    # detection -- it must follow the switch, not keep reporting the first
+    # track's clap.
+    mpv.aid = 2
+    qapp.processEvents()
+    time.sleep(0.2)
+    switched_index = win._current_audio_stream_index()
+    print(f"resolved ff-index after switching to aid=2: {switched_index}")
+    assert switched_index == 2, f"expected ff-index 2 after switching tracks, got {switched_index}"
+
+    win.source_panel.marker_strip.clear()
+    win.detect_all_button.click()
+    qapp.processEvents()
+    ok = wait_for(lambda: win.detect_all_button.isEnabled(), qapp)
+    assert ok, "detection did not finish in time"
+    markers = win.source_panel.marker_strip.markers()
+    print(f"markers with second track selected: {markers}")
+    assert any(abs(m - 7.0) < 0.5 for m in markers), f"expected a marker near 7.0s (second track's clap) after switching: {markers}"
+    assert not any(abs(m - 3.0) < 0.5 for m in markers), f"should not see the FIRST track's 3.0s clap after switching: {markers}"
+
+    win.close()
+    qapp.processEvents()
+    print("\nPASS: clap detection follows mpv's actual selected audio track, not ffmpeg's default")
+
+
+if __name__ == "__main__":
+    main()