perf: Use Charset in StringUtils to/fromUtf8.

[gianm]

There is an old optimization in StringUtils#fromUtf8 and toUtf8 that has outlived its usefulness. Apparently it used to be faster to pass the String charset name rather than a Charset object, but that is no longer the case.

[gianm]

Benchmarks. Not checked in, they're just doing the two alternatives on various lengths of strings.

Benchmark                                    (length)  Mode  Cnt     Score    Error  Units
Utf8CharsetBenchmark.fromUtf8_charsetName           8  avgt   10     9.017 ±  0.040  ns/op
Utf8CharsetBenchmark.fromUtf8_charsetName          64  avgt   10    12.426 ±  0.160  ns/op
Utf8CharsetBenchmark.fromUtf8_charsetName        1024  avgt   10    43.674 ±  0.298  ns/op

Utf8CharsetBenchmark.fromUtf8_charsetObject         8  avgt   10     6.724 ±  0.079  ns/op
Utf8CharsetBenchmark.fromUtf8_charsetObject        64  avgt   10    10.231 ±  0.136  ns/op
Utf8CharsetBenchmark.fromUtf8_charsetObject      1024  avgt   10    41.794 ±  0.448  ns/op

Utf8CharsetBenchmark.toUtf8_charsetName             8  avgt   10    10.255 ±  1.114  ns/op
Utf8CharsetBenchmark.toUtf8_charsetName            64  avgt   10    11.069 ±  0.279  ns/op
Utf8CharsetBenchmark.toUtf8_charsetName          1024  avgt   10    42.732 ±  0.385  ns/op

Utf8CharsetBenchmark.toUtf8_charsetObject           8  avgt   10     3.812 ±  0.063  ns/op
Utf8CharsetBenchmark.toUtf8_charsetObject          64  avgt   10     7.024 ±  0.045  ns/op
Utf8CharsetBenchmark.toUtf8_charsetObject        1024  avgt   10    39.115 ±  0.431  ns/op

[FrankChen021]

🟢 Approval recommended

No actionable issues found in this review. The production change keeps the UTF-8 conversion behavior while using StandardCharsets.UTF_8 directly, and the repository-wide references, surrounding call sites, compatibility guidance, and all changed tests were inspected.

Reviewed 6 of 6 changed files.

---

This is an automated review by Codex GPT-5.6-Luna(max)