[PR] Adds rare antag token drops for low population servers

[Piesuu]

I feel like there should be more edge case preventions in the code (too low of a pop in general, for example) as well as making sure antag tokens can't just make it so that half the server rolls antag, unless that's already a thing.

The only idea I have in mind is separating admin-given Antag Tokens and end-round given Antag Tokens into separate categories, where the end-round token Doesn't guarantee a roll if there are enough antags in play already for the server pop.

[Danger Noodle]

Personally I'm in favor of giving it a shot, it certainly sounds interesting. A few guardrails around antag count might need to be put up but it doesn't sound all that likely to actually end up as a big issue.

[JORJ949]

On what Snoid said about antag being easier to get on RP: This is not the case, while there are more people with antag disabled, there are also less antags in general (to the point i often go "what the fuck 1 ling 1 sleeper at 60 pop?" and the rounds are longer, end result being I feel like I get antag about as often as I did on classic, that being rare but not insanely rare.

Antag tokens will likely not fix this as this is caused by a significant amount of people joining in a single round, what would actually fix this is scaling latejoin antagonist chances by the amount of already existing antagonists(dead or alive) in comparison to the gamemode's enemy rate, so if you've one traitor on 60 pop when the gamemode expects 4 or 5 there'll be a relatively high chance a latejoiner rolls antag.

Like I commented on the PR I'd also like to see more things taken into account such as how long the round lasted and the actual population difference, calling the shuttle over a minor inconvenience is already a bit of a problem on classic, having it so that ending rounds earlier means more antagging will exacerbate this issue, in regards to the actual population difference 3 vs 10 pop is IMO a fine popgap that doesn't need tokens, but the current system will go "wow there are over triple the players on the other server, better hand out some tokens at a relatively higher rate!"

[Johnny Faith]

This would probably help classic at least a little bit with player count, or at least the 10 of us that still play classic. I'd be happy to see what happens.

[NightmarechaMillian]

having been on both sides of the high/lowpop coin I'm definitely a big fan of this. 

Spend your goddamn tokens omg (hopefully people will be more likely to if they're less Unique and Magical anyway)

I do feel like this should be telegraphed more, or perhaps changed. prior to this PR I've gotten (after running the numbers 0.25 antag tokens per Real Life Year of playing regularly, which is considerably Not A Lot™. Throwing any nuance out the window and assuming a base chance of 20%, you can get about 1 a week, 2 if you play more than 1 round. That's kind of a crazy jump and I feel like it's worth conveying "hey, this won't increase your antag tokens above 1". There is the bold text saying so to the player, but given the sheer chaos of shuttle rides and EROG there's absolutely some other signal needed about the intricacies of the mechanic. It's definitely better than reducing the chance (grinding for ingame rewards, ehh), but if the testmerge goes through, maybe it'd be worth letting that number go up to 3? Unsure what the downsides would be.

As far as my experience in Will This Make RP/Lowpop/Goonstation/Wire's house Literally Blow Up question, I'll be honest and admit I have not noticed a single difference. RP seems to largely be RP and lowpop seems to be slightly higher lowpop, sometimes maybe if you get on at the right times. AFAIC this seems to be the carrot on a stick needed to push the servers towards a slightly better equillibrium.