dr mario env

[doofenshirmtz]

rewards configs were modified but perfs dont exist save me jospeh

[jsuarez5341]

Solid first PR, reviewed on stream. Minor style stuff:

• don't newline curlybraces
• Learn inverted if pattern / early return
• Don't 1-line continue statements
• Semantic grouping for ops (a + b*c not a + b * c)
• When you get really long conditionals because the vars have long names, rename or temp var to shorten

I will help debug training on stream. PR some small fixes with the above and then yourself + collaborator to the contributors list on the website repo. Have him add that dino env too, I missed that one!