Add support for language in Linux platform voices

This CL is keeping track of the language returned by the speak
synthetizer. The language attribute is used by the TTS controller
GetMatchingVoice(...) to choose the most suitable voices.


For an empty test that do:
  speechSyntheizer.speak(...)

My local computer is choosing:
  Afrikaans espeak-ng
which is the first alphabetical entry.

With the patch, it is choosing:
  English (America) espeak-ng
since my app-locale are taken into account by the TTS controller.

Bug: 1133813
Change-Id: Ie04657dec7f947ddd91c9e672bcb36a54863d88e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2493021Reviewed-by: Dominic Mazzoni <dmazzoni@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Commit-Queue: Etienne Bergeron <etienneb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#820324}

content/browser/speech/tts_linux.cc

@@ -33,6 +33,7 @@ namespace {
 struct SPDChromeVoice {
   std::string name;
   std::string module;
+  std::string language;
 };
 
 using PlatformVoices = std::map<std::string, SPDChromeVoice>;
@@ -264,6 +265,7 @@ void TtsPlatformImplBackgroundWorker::InitializeVoices(PlatformVoices* voices) {
       SPDChromeVoice spd_data;
       spd_data.name = spd_voice->name;
       spd_data.module = module;
+      spd_data.language = spd_voice->language;
       std::string key;
       key.append(spd_data.name);
       key.append(" ");
@@ -488,6 +490,7 @@ void TtsPlatformImplLinux::GetVoices(std::vector<VoiceData>* out_voices) {
     VoiceData& voice = out_voices->back();
     voice.native = true;
     voice.name = it->first;
+    voice.lang = it->second.language;
     voice.events.insert(TTS_EVENT_START);
     voice.events.insert(TTS_EVENT_END);
     voice.events.insert(TTS_EVENT_CANCELLED);