fix(tutorial): correct return order of min/max ms in grouped_gemm benchmark

[niyunsheng]

New contributor declaration

• I am not making a trivial change, such as fixing a typo in a comment.

• I have written a PR description following these
  rules.

• I have run pre-commit run --from-ref origin/main --to-ref HEAD.

• Select one of the following.

  • I have added tests.
    • /test for lit tests
    • /unittest for C++ tests
    • /python/test for end-to-end tests
  • This PR does not need a test because FILL THIS IN.

• Select one of the following.

  • I have not added any lit tests.
  • The lit tests I have added follow these best practices,
    including the "tests should be minimal" section. (Usually running Python code
    and using the instructions it generates is not minimal.)

Description

This PR fixes an issue with inverted error bars in the grouped_gemm tutorial's performance plot.

Currently, the benchmark_batches and benchmark_square_matrices function returns ms, max_ms, min_ms. Since the y-axis metric is absolute runtime(ms), this inadvertently swaps the upper and lower bounds for the error bars in the triton.testing.perf_report rendering logic.

(Note: Other tutorials correctly use the inverted order because their y-axis is throughput, e.g., gbps(min_ms) as the upper bound. This issue is isolated to grouped_gemm.)