Add some nicer formatting

Add fast ecvt, fcvt and pretty fmtUncertain

Minor speed-up of compilation time

For some cases like rp.

Hot bug fixes

parseInt didn't do negatives; parseFloat was quite hosed. All better now and with a test suite.

Quick rename of rx -> rp

Avoid regular expression abbreviation connotations.

Fix Nim version portabilty of mslice

Sorry for the slight 10 vs 10'u64 hiccup.

`rx`, Fast MSlice number parsers, fmtUncertainMerged

Add fmtUncertainMerged - The Particle Data Group's neatly terse uncertainty notation { 12.34(56) = 12.34 +- 0.56 - 2 significant figures in the error and aligned digits.}. This scales to high & low exponents quite well. { One might argue a '/' could be clearer & shorter as in 12.34/56 but this is non-standard. }

Add fast MSlice number parsers for int & float. These do not need to first convert to a Nim string.

Add rx - a row executor/row processer program generator that allows a work flow like "awk one-liners" but for Nim programs..So, a delay to compile, but then static typing & much better performance on meso-scale data like a few 100 million records.

Tweak strUt/joins & add conncomp

Add a connected component finder. Also simplify one of the two strUt/joins.

Add cligen/prints module with `[fe]print[n]`

Got sick of hearing about it over at https://forum.nim-lang.org/t/7819

So, I wrote it in a few minutes. It probably has flaws, but also nothing is perfect.

Add fmtUncertain

Add some utility code to cligen/strUt to suppress meaningless digits "the smart way" (or at least the way particle physicists have been doing it for >30 years).

Make rst/md presence more amenable to help2man

Both start & stop ANSI SGR sequences can be included in all the emitted help. This can make it easier to automate translation to troff for man page contexts. Also add a few new osUt utility APIs.