Improves long-significand performance

[lemire]

If you have binary floating-point numbers, you ought to serialize them using no more than 17 digits. Anymore and you are being wasteful. Our code is optimized for such a scenario. However, if someone has data with lots and lots of digits per floating-point number, then our performance is somewhat less. This PR is better if you expect to be processing long significand.

Configuration: AMD Rome Processor (GCC 10)

model	before	after (this PR)	gain/loss
uniform	1132.06 MB/s	1121.70 MB/s	-1%
uniform/concise	951.12 MB/s	952.04 MB/s	0%
canada	802.83 MB/s	824.92 MB/s	+3%
big_ints	370.35 MB/s	520.40 MB/s	+ 40%

See also #27 for a different trade-off.