improve performance with parsing bigints with lots of digits

[pjfanning]

Parsing numbers in Java has poor performance characteristics. Parse time increases exponentially as the number of digits increases.
Using BigDecimalParser to parse BigIntegers has better performance when there are a lot of digits. It is slower for more usual numbers. From benchmarking, BigDecimalParser is more performant when you have approx 1250 digits and as you head to 10k, 100k digits, it is much faster. BigDecimalParser is not O(n) but it is better than new BigInteger(string).

Relates to #814

[cowtowncoder]

Interesting! Ok, I trust that you have benchmarked this in addition to it being found a useful optimization in other libraries.
Will merge to go on 2.14.0 since it seems like a low-risk improvement.

[pjfanning]

@cowtowncoder I benchmarked using https://github.com/pjfanning/jackson-number-parse-bench

On my laptop, I got these results

Benchmark                                 Mode  Cnt       Score       Error  Units
BigIntegerParserBench.bigDec1000         thrpt    5   35195.566 ±  2758.478  ops/s
BigIntegerParserBench.bigDec10000        thrpt    5    1232.178 ±   221.485  ops/s
BigIntegerParserBench.bigDec1000000      thrpt    5       0.396 ±     0.031  ops/s
BigIntegerParserBench.bigInt1000         thrpt    5   48308.680 ±   671.193  ops/s
BigIntegerParserBench.bigInt10000        thrpt    5     617.721 ±    14.032  ops/s
BigIntegerParserBench.bigInt1000000      thrpt    5       0.063 ±     0.002  ops/s
BigIntegerParserBench.fastBigDec1000     thrpt    5  138409.766 ± 10781.948  ops/s
BigIntegerParserBench.fastBigDec1000000  thrpt    5       2.342 ±     0.019  ops/s

• the higher ops/s is the better
• 1000 in benchmark means testing with integers with 1000 chars, 10000 means 10000 chars, etc.
• fastBigDec benchmarks use an experimental new parser that is not ready for Jackson 2.14
• bigDec benchmarks use BigDecimalParser
• bigInt benchmarks use new BigInteger(string)

[cowtowncoder]

@pjfanning Cool. Would it make sense to add simple results on README (everyone can of course run it locally too).