Mumble tweaks/fixes

[AvarianKnight]

Goal of this PR

Fix some possible edge cases where players could be invalid but we would still send a properly formatted mumble name

Removes swapping between std::wstring and std::string

Reset the mumble connection in cases where the client loses their TCP connection.

Fix some cases where the client would send duplicates of the same targets to the server, leading to some excessively large packets.

Wrap any native that needs to be used while connected in MakeMumbleNative

This change set is currently untested

This PR applies to the following area(s)

FiveM, RedM

Successfully tested on

Game builds: ..

Checklist

• Code compiles and has been tested successfully.
• Code explains itself well and/or is documented.
• My commit message explains what the changes do and what they are for.
• No extra compilation warnings are added by these changes.

[AvarianKnight]

This should be ready to be reviewed now, newer changes remove conversions between string types and resets the players mumble connection if the server doesn't respond to TCP pings.

[Nobelium-cfx]

Looks like a really nice refactoring. Thank you!

Overall LGTM, just left some style related comments and asked some clarifying questions. PTAL.

It is a big change tho, so before merging we would probably need to tests it on some server.

[Nobelium-cfx]

Nit: this seems to be very similar to getAudioContext. Can we use getMumbleName in getAudioContext to avoid duplication?

[AvarianKnight]

iirc this was actually needed, getAudioContext doesn't check for if they're connected before hand, while everywhere that uses getMumbleName does, we can just do the connected check again if wanted though.

[Nobelium-cfx]

What I meant is that it seems that we could refactor getAudioContext as below to reduce some duplication.

static std::shared_ptr<lab::AudioContext> getAudioContext(int playerId)
{
        auto name = getMumbleName(playerId);
	if (name  == nullptr || !IsMumbleConnected())
	{
		return {};
	}
	return g_mumbleClient->GetAudioContext(name);
}