LeetCUDA v3.0.6

What's Changed

• misc: update merge_attn_states unit tests by @DefTruth in #281
• misc: update merge_attn_states docs by @DefTruth in #282
• misc: update merge_attn_states docs by @DefTruth in #283
• feat: remove merge_attn_states kernel help func by @DefTruth in #284
• misc: remove static flag for to/from_float by @DefTruth in #285
• misc: add new zhihu tech blog link by @DefTruth in #287
• misc: add debug flag for ncu profile by @DefTruth in #288
• bugfix: corrected theta calculation in RoPE CUDA kernel by @jiaau in #290
• docs: Add my ring-attention zhihu blog by @DefTruth in #291
• Add simple CuTe mat-transpose implementations by @botbw in #292
• Update README.md by @DefTruth in #296
• Update README.md by @DefTruth in #297
• Update README.md by @DefTruth in #298
• Rename to LeetCUDA by @DefTruth in #299

New Contributors

• @jiaau made their first contribution in #290
• @botbw made their first contribution in #292

Full Changelog: v3.0.5...v3.0.6