Formulation of reverse diffusion process in DDPM ( 1 - alpha_prod_t = beta_prod_t assumption)

### Describe the bug

I looked into sampling code of DDPM, and I believe there's a mistake:
#### I believe the code makes assumption that 1 - alpha_prod_t = beta_prod_t, which simply isn't true.

Original sampling algorithm:
[original paper](https://arxiv.org/pdf/2006.11239)

x_0 prediction from sample and predicted noise :

Eqn 15 from paper, as referenced in code snippet as well:

code implementation in diffusers:
https://github.com/huggingface/diffusers/blob/1e8cf2763d37fa66cfba2fb87acfe3443068ce43/src/diffusers/schedulers/scheduling_ddpm.py#L446-L449

### Here, it clearly seems like the 1 - alpha_prod_t = beta_prod_t is being used.

investigated from post by @AlejandroBaron in https://github.com/huggingface/diffusers/discussions/9431_

### Reproduction

I haven't tested out code, but this seems fundamental formulation issue.

### Logs

_No response_

### System Info

0.30.3

### Who can help?

@sayakpaul @DN6 @yiy

Start in src/diffusers/schedulers/scheduling_ddpm.py at lines 446-449, then compare the implementation with Eq. 15 of the linked DDPM paper and the referenced discussion. Verify whether the reported relationship between alpha_prod_t and beta_prod_t is actually assumed, and establish the correct behavior with a reproducible check before proposing a fix.

Vom Indexierungsmodell aus dem Issue-Text verfasst.