reduce participation points in match_costs

[chezbananas]

weight participation in matches less, focusing more on raw score
maybe begin logistically lowering scores for people who played less than 25% of maps or something idk lol
this is especially noticeable in stuff like this match:
https://osu.ppy.sh/community/matches/53275090
seriously how the fuck is corim rated below me lol
idk i might be overlooking something here but just like fix it lol

[dain98]

I nerfed constant k in the formula from 0.4 to 0.2, the values go from:
Blue Team:
1: DigitalHypno - 1.94
2: Cappy - 1.85
3: Skill - 1.44
4: Fish- - 1.38
5: Boggles - 1.35
6: Exarch - 1.27

Red Team:
1: Toy - 1.94
2: fieryrage - 1.80
3: chezbananas - 1.56
4: Tohru - 1.46
5: Corim - 1.28
6: byfar - 0.77

to

Blue Team:
1: DigitalHypno - 1.94
2: Cappy - 1.88
3: Skill - 1.47
4: Fish- - 1.43
5: Boggles - 1.39
6: Exarch - 1.31

Red Team:
1: Toy - 1.95
2: fieryrage - 1.81
3: chezbananas - 1.60
4: Tohru - 1.50
5: Corim - 1.31
6: byfar - 0.80

I guess it's a bit more accurate now, but the issue here is that each point cost depends on how well the player performed compared to the rest of the lobby. If you look through Corim's points, he performed well but compared to others, the bot detects it as mediocre performance. I can't really do much about that unless I completely rework the formula. Again.

[dain98]

I'll keep this issue open in case someone comes up with a good solution.