unipc schedule throws exception when use_karras_sigmas=True

[reimager]

Describe the bug

unipc scheduler throws exception when use_karras_sigmas=True

File "/usr/local/src/diffusers/src/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py", line 642, in call
self.scheduler.set_timesteps(num_inference_steps, device=device)
File "/usr/local/src/diffusers/src/diffusers/schedulers/scheduling_unipc_multistep.py", line 239, in set_timesteps
sigmas = self._convert_to_karras(in_sigmas=sigmas, num_inference_steps=num_inference_steps)
File "/usr/local/src/diffusers/src/diffusers/configuration_utils.py", line 137, in getattr
raise AttributeError(f"'{type(self).name}' object has no attribute '{name}'")
AttributeError: 'UniPCMultistepScheduler' object has no attribute '_convert_to_karras'

Its just missing the _convert_to_karras and _sigma_to_t functions.
I will submit a pull request shortly

Reproduction

specify use_karras_sigmas=True when creating unipc scheduler, then use scheduler

Logs

No response

System Info

• diffusers version: 0.19.3
• Platform: Linux-5.10.0-23-amd64-x86_64-with-glibc2.35
• Python version: 3.10.6
• PyTorch version (GPU?): 2.1.0a0+b5021ba (True)
• Huggingface_hub version: 0.16.4
• Transformers version: 4.32.0.dev0
• Accelerate version: 0.22.0.dev0
• xFormers version: 0.0.21+db0e439.d20230810
• Using GPU in script?: yes
• Using distributed or parallel set-up in script?: no

Who can help?

No response