feat(etl): 实现 LRC 时间戳去除

+import re
+
+_TIMESTAMP_RE = re.compile(r'\[\d{2}:\d{2}\.\d{2,3}\]')
+_META_TAG_RE = re.compile(r'\[[a-zA-Z]+:[^\]]*\]')
+
+
+def strip_timestamps(text: str | None) -> str:
+    if not text:
+        return ''
+    text = _META_TAG_RE.sub('', text)
+    text = _TIMESTAMP_RE.sub('', text)
+    lines = [line.strip() for line in text.splitlines() if line.strip()]
+    return '\n'.join(lines)

+from etl_to_crawler.lyric import strip_timestamps
+
+def test_strips_standard_timestamps():
+    lrc = "[00:12.34]第一行歌词\n[01:23.45]第二行歌词"
+    assert strip_timestamps(lrc) == "第一行歌词\n第二行歌词"
+
+def test_strips_millisecond_timestamps():
+    lrc = "[00:12.345]带三位毫秒"
+    assert strip_timestamps(lrc) == "带三位毫秒"
+
+def test_strips_meta_tags():
+    lrc = "[ti:歌曲名]\n[ar:歌手]\n[00:01.00]歌词"
+    result = strip_timestamps(lrc)
+    assert "歌词" in result
+    assert "[ti:" not in result
+
+def test_empty_and_none():
+    assert strip_timestamps(None) == ''
+    assert strip_timestamps('') == ''
+
+def test_plain_text_unchanged():
+    assert strip_timestamps("没有时间戳的歌词") == "没有时间戳的歌词"