Update decode kernel benchmark with new Triton backend interface

[amosyou]

Motivation

The Triton kernel for decode attention was updated with a new backend interface in #3292, breaking the benchmark code.

Modifications

Corrected the import for should_use_tensor_core and replaced req_to_token, b_req_idx, b_seq_len with kv_indptr and kv_indices.

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://slack.sglang.ai to discuss your PR.

[amosyou]

cc @zhyncs

[amosyou]

hi @zhyncs ! this should be good to go, let me know if there's any concern

[github-actions]

This pull request has been automatically closed due to inactivity. Please feel free to reopen it if needed.