Refactor and release Transformers patching and auto-kernelize loop & other updates

[hannahli-nv]

Description

Update codes.

This PR contains 10 new commit(s).

Commits included:

081005d Add cutile-autotuning skill
7192508 [skill] update improve-cutile-perf skill
a2c8a8d [skill] refine monkey-patch-kernels-to-transformers based on feedback
87d29d5 skills: Drop HTML-comment headers from non-SKILL.md markdown files
970478e Batch cuTile cat copies
2d4cf0f Fix per-call replace_hints JIT-cache invalidation in decode loops
d58a3a5 [benchmark] add DUMP_CUPTI_EVENTS dump in benchmark_fn_cupti
fba356e bump version to 1.3.0
3fa4469 test: remove xfail marker for test_rope_quantize_fp8.py
7ba2266 Refactor and release Transformers patching and auto-kernelize loop

CI Configuration

config:
  build: true
  # valid options are "ops" and "benchmark"
  test: ["ops", "benchmark"]

Checklist

• Code formatted and imports sorted via repo specifications (./format.sh)
• Documentation updated (if needed)
• CI configuration reviewed

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[hannahli-nv]

/ok to test a2c8a8d

[hannahli-nv]

/ok to test 081005d