Use cooperative-groups instead of cub warp-reduce for strings contains

[davidwendt]

Description

Replaces the cub::WarpReduce usage in cudf::strings::contains with cooperative-groups any().
The change is only for the contains_warp_parallel kernel which is used for wider strings.
Using cooperative-groups generates more efficient code for the same results and gives an additional 11-14% performance improvement.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[davidwendt]

Benchmark results show no change for smaller strings as expected but an 11-14% improvement for wider strings:

## [0] NVIDIA RTX 5880 Ada Generation

| max_width | num_rows | hit_rate |   api    |  target  |   Ref Time |   Cmp Time |        Diff |   %Diff |
|-----------|----------|----------|----------|----------|------------|------------|-------------|---------|
|    32     |  32768   |    20    | contains |  scalar  |  38.918 us |  38.654 us |   -0.264 us |  -0.68% |
|    64     |  32768   |    20    | contains |  scalar  |  64.614 us |  64.554 us |   -0.059 us |  -0.09% |
|    128    |  32768   |    20    | contains |  scalar  |  53.720 us |  49.504 us |   -4.217 us |  -7.85% |
|    256    |  32768   |    20    | contains |  scalar  |  68.812 us |  63.345 us |   -5.467 us |  -7.95% |
|    32     |  262144  |    20    | contains |  scalar  |  50.448 us |  50.143 us |   -0.304 us |  -0.60% |
|    64     |  262144  |    20    | contains |  scalar  |  91.182 us |  90.712 us |   -0.470 us |  -0.52% |
|    128    |  262144  |    20    | contains |  scalar  | 247.065 us | 214.438 us |  -32.627 us | -13.21% |
|    256    |  262144  |    20    | contains |  scalar  | 355.056 us | 313.134 us |  -41.922 us | -11.81% |
|    32     | 2097152  |    20    | contains |  scalar  | 175.870 us | 174.549 us |   -1.321 us |  -0.75% |
|    64     | 2097152  |    20    | contains |  scalar  | 427.138 us | 426.376 us |   -0.763 us |  -0.18% |
|    128    | 2097152  |    20    | contains |  scalar  |   1.791 ms |   1.539 ms | -252.440 us | -14.09% |
|    256    | 2097152  |    20    | contains |  scalar  |   2.638 ms |   2.308 ms | -330.054 us | -12.51% |
|    32     |  32768   |    80    | contains |  scalar  |  37.129 us |  37.174 us |    0.045 us |   0.12% |
|    64     |  32768   |    80    | contains |  scalar  |  63.504 us |  63.180 us |   -0.324 us |  -0.51% |
|    128    |  32768   |    80    | contains |  scalar  |  57.137 us |  52.307 us |   -4.831 us |  -8.45% |
|    256    |  32768   |    80    | contains |  scalar  |  72.910 us |  66.125 us |   -6.786 us |  -9.31% |
|    32     |  262144  |    80    | contains |  scalar  |  53.414 us |  53.214 us |   -0.200 us |  -0.37% |
|    64     |  262144  |    80    | contains |  scalar  |  98.814 us |  98.445 us |   -0.370 us |  -0.37% |
|    128    |  262144  |    80    | contains |  scalar  | 277.457 us | 239.903 us |  -37.554 us | -13.54% |
|    256    |  262144  |    80    | contains |  scalar  | 385.274 us | 334.618 us |  -50.656 us | -13.15% |
|    32     | 2097152  |    80    | contains |  scalar  | 210.464 us | 208.620 us |   -1.845 us |  -0.88% |
|    64     | 2097152  |    80    | contains |  scalar  | 412.716 us | 411.962 us |   -0.754 us |  -0.18% |
|    128    | 2097152  |    80    | contains |  scalar  |   2.032 ms |   1.743 ms | -289.033 us | -14.22% |
|    256    | 2097152  |    80    | contains |  scalar  |   2.875 ms |   2.477 ms | -397.693 us | -13.83% |

[PointKernel]

Loos great! 🔥

[davidwendt]

/merge