Add chars-tokenizer to nvtext tokenize_benchmark.cpp

[davidwendt]

This PR adds a benchmark to the current tokenize_benchmark.cpp to measure the nvtext::character_tokenize API.

PR #8085 added code for using the nvtext::character_tokenize function.
The benchmark was also useful while investigating #8094.
Also found and removed an unused variable in the code logic.

[codecov]

Codecov Report

Merging #8125 (10537f7) into branch-0.20 (51336df) will decrease coverage by 0.36%.
The diff coverage is 90.90%.

❗ Current head 10537f7 differs from pull request most recent head c6ca384. Consider uploading reports for the commit c6ca384 to get more accurate results

@@               Coverage Diff               @@
##           branch-0.20    #8125      +/-   ##
===============================================
- Coverage        82.88%   82.52%   -0.37%     
===============================================
  Files              103      103              
  Lines            17668    17500     -168     
===============================================
- Hits             14645    14441     -204     
- Misses            3023     3059      +36     

Impacted Files	Coverage Δ
python/cudf/cudf/core/column/__init__.py	100.00% <ø> (ø)
python/cudf/cudf/core/groupby/groupby.py	91.39% <ø> (-0.06%)	⬇️
python/cudf/cudf/core/index.py	92.62% <ø> (-0.46%)	⬇️
python/cudf/cudf/core/indexing.py	96.50% <ø> (+0.21%)	⬆️
python/cudf/cudf/core/reshape.py	91.08% <ø> (-0.03%)	⬇️
python/cudf/cudf/core/scalar.py	87.87% <ø> (+0.13%)	⬆️
python/cudf/cudf/core/series.py	91.40% <ø> (-0.32%)	⬇️
python/cudf/cudf/core/tools/datetimes.py	80.17% <ø> (-4.36%)	⬇️
python/cudf/cudf/core/window/rolling.py	88.88% <ø> (+0.65%)	⬆️
python/cudf/cudf/io/orc.py	86.80% <ø> (-0.10%)	⬇️
... and 71 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f686c01...c6ca384. Read the comment docs.

[harrism]

@gpucibot merge