arrow

ARROW-9710: [C++] Improve performance of Decimal128::ToString by 10x, and make the implementation reusable for Decimal256.

Added a benchmark to decimal_benchmark.

```
Running release/arrow-decimal-benchmark
Run on (12 X 4500 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x6)
  L1 Instruction 32 KiB (x6)
  L2 Unified 1024 KiB (x6)
  L3 Unified 8448 KiB (x1)
Load Average: 0.23, 0.22, 0.23
----------------------------------------------------------------------------------
Benchmark                        Time             CPU   Iterations UserCounters...
----------------------------------------------------------------------------------
ToString (before)             2205 ns         2205 ns       321603 items_per_second=2.72094M/s
ToString (after)               216 ns          216 ns      3235082 items_per_second=27.7865M/s
```

Closes #7945 from MingyuZhong/decimal_to_string

Lead-authored-by: Mingyu Zhong <myzhong@google.com>
Co-authored-by: Antoine Pitrou <antoine@python.org>
Signed-off-by: Antoine Pitrou <antoine@python.org>