More specifically:

Add a flavor of mUsed for `scrollsy` - total-complement to mAvl: `mUAv`

Right align headers for right aligned fields

Rename NI -> NICE; F -> FLAGS; Abbreviation unhelpful

TTY -> TT & 2-column format to save a terminal column in the common case.

In more detail:
- Add new parts per billion CPU time usage as %b; Update configs/cb0/style.auser to use it.
- Give 2 new ways to test for existence of any matches:
- pf -e STUFF || start-it
- pf -ac STUFF >/dev/null || start-it
Fix:
- bug converting delay TimeSpec from seconds to jiffies
- Fix longstanding environ 'e' masks %cPU 'e' bug

..where it should perhaps always have been.

Add proportional set size (PSS) reporting from Linux /proc/PID/smaps file.

Better bug fix for differential reports with differences < 0 (as happens a lot with PSS, actually, due to sharing dynamics). procs d -d1 really does seem nicer than top to me..It's certainly more general.

Correct the nowNs field width.

Together this lets you do something like this to monitor a process:

burn-cpu & pd -pBHd1 -D'R<>J' -f "%N %R %< %> %J %a" $!

Or you could replace $! with $(pf firefox), take out the -p to leave colors on, or etc., etc.

This is more a hygiene bump since while procs is Linux-only and Linux has always provided /proc/PID/exe , but procs.nim does call that updated API code.

Also some documentation work and a dep update.

The one command install like lc is maybe the most visible thing.

Run a mini-install with mini= nimble install procs to avoid that fanciness provided for more "just try it out" users.

Also in this release:

• working under mm:arc by working around a bug
• a minor new column in default config
• a somewhat faster procs find

This is still a work in progress. (I should, at least, write 3 man pages and install the config from nimble as lc does, and perhaps even optionally install symbolic links for pd/pf/pk/pw/sc out of nimble for folks doing quick try-outs.)

Should work as intended now. Stiff iffy-tie breaking if PID wraparound can happen, but then again how many ties within the 10ms jiffy are likely?