Improve IEC binary and SI decimal prefix handling

Improve IEC binary and SI decimal prefix handling

Use kb_base=1000 to follow international standards for unit prefixes.

To specify power-of-10 decimal values defined in the International
System of Units (SI):
	Ki means kilo (K) or 1000
	Mi means mega (M) or 1000**2
	Gi means giga (G) or 1000**3
	Ti means tera (T) or 1000**4
	Pi means peta (P) or 1000**5

To specify power-of-2 binary values defined in IEC 80000-13:
	k means kibi (Ki) or 1024
	M means mebi (Mi) or 1024**2
	G means gibi (Gi) or 1024**3
	T means tebi (Ti) or 1024**4
	P means pebi (Pi) or 1024**5

For example, this specifies a blocksize of 4096 bytes:
	kb_base=1000
	bs=4KiB

With kb_base=1024 (the default), the unit prefixes are opposite from
those specified in the SI and IEC 80000-13 standards to provide
compatibility with old scripts.  For example, this specifies a
blocksize of 4096 bytes:
	kb_base=1024
	bs=4K

For outputs printing quantities and bandwidths:
* eta stats only use the preferred prefix
* final stats include both (non-preferred prefix in parenthesis)
* in gfio, all windows include both

Text outputs are rearranged to try to obviously break any scripts
parsing the output rather than silently confuse them.  The terse
and json outputs, which are intended for parsing, are unchanged.

Old:
Jobs: 576 (f=576), CR=86.4GB/576MB KB/s: [w(288),r(288)] [0.0% done] [33884MB/27114MB/0KB /s] [8471K/6778K/0 iops] [eta 06h:59m:57s]
   read: io=363338MB, bw=34014MB/s, iops=8304.3K, runt= 10682msec
  write: io=282447MB, bw=26225MB/s, iops=6402.7K, runt= 10770msec
   READ: io=363338MB, aggrb=34014MB/s, minb=34014MB/s, maxb=34014MB/s, mint=10682msec, maxt=10682msec
  WRITE: io=282447MB, aggrb=26225MB/s, minb=26225MB/s, maxb=26225MB/s, mint=10770msec, maxt=10770msec

New:
Jobs: 576 (f=576), 590MB/s-89.2GB/s: [w(288),r(288)][0.0%][r=34.2GB/s,w=26.3GB/s][r=8542k,w=6572k IOPS][eta 06h:59m:55s]
   read: IOPS=8362k, BW=34.3GB/s (31.1GiB/s)(665GB/19417msec)
  write: IOPS=6423k, BW=26.4GB/s (24.6GiB/s)(511GB/19431msec)
   READ: bw=34.3GB/s (31.1GiB/s), 34.3GB/s-34.3GB/s (31.1GiB/s-31.1GiB/s), io=665GB (619GiB), run=19417-19417msec
  WRITE: bw=26.4GB/s (24.6GiB/s), 26.4GB/s-26.4GB/s (24.6GiB/s-24.6GiB/s), io=511GB (476GiB), run=19431-19431msec

Documentation changes are in a subsequent patch.

Signed-off-by: Jens Axboe <axboe@fb.com>