(TTML) Add an optional way to separate speakers by style

[vevv]

Some subtitles use different styles for different speakers, those could be converted to dashes.

<p begin="00:02:04.120" end="00:02:07.640" region="agr_0.5_20_39.5_4_after" style="S2" xml:id="C30_1">
  <span style="S3">Someone get that, will you?</span>
  <br/>
  <span style="S4">Liam's turn.</span>
</p>

<p begin="00:07:32.280" end="00:07:35.160" region="agr_0.5_20_39.5_4_after" style="S2" xml:id="C125_1">
  <span style="S5">Are you gonna do it again before</span>
  <br/>
  <span style="S5">school?</span>
  <span style="S3">Give him a break,</span>
</p>

The fix for hanging single-words will have to take dashes into account, as currently it processes the second example as:

122
00:07:32,280 --> 00:07:35,160
Are you gonna do it again before
school? Give him a break,

Where the ideal probably is:

122
00:07:32,280 --> 00:07:35,160
- Are you gonna do it again before school?
- Give him a break,

Unfortunately this is still not ideal, but probably the best that can be done in SRT, as those are originally positional subs.