Fixed multiple particle issues: division by zero, color ramp override, scale dependent on amount ratio

[QbieShay]

Closes #83258
Fixes scale and amount ratio being tied together
Fixes initial ramp overriding color

The last two issues were reported to me in DM

[Ansraer]

Discussed during the meeting today, we have concerns that the current random function generates numbers in a repeatable pattern. This is something we would ideally like to investigate before merging.

EDIT: I am going to brute force test the rand_from_seed function after the meeting.
EDIT 2: Ok, it would appear that rand_from_seed does properly deliver semi random numbers. I generated several thousand values and couldn't find any pattern. For now, it would appear that we need to look for the problem somewhere else.
Clay figured it out while we were talking about it, your solution is correct.

[clayjohn]

Discussed in this week's rendering meeting. Everything looks good. The concern I had over moving the AMOUNT_RATIO check turned out to be unfounded. Moving AMOUNT_RATIO is the correct fix

[akien-mga]

Thanks!