Aiter fp8 kv cache

[kkHuang-amd]

Motivation

Support fp8 kv cache in aiter-backend of AMD.

Aiter backend only support mla decode fp8 computation.

Other attention function still do bf16 computation

Modifications

Aiter backend and model runner.

Next actions

1. Support other attention function also do fp8 computation
2. Accuracy issue in MTP run without SGLANG_AITER_MLA_PERSIST

Accuracy Tests

Benchmarking and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.