Add optional substring matching for picker

[pascalkuthe]

This PR implements common syntax that allows matching substrings instead of fuzzy matching. I will quote the fzf README here for an explanation what exactly this does:

Token	Match type	Description
sbtrkt	fuzzy-match	Items that match sbtrkt
'wild	exact-match (quoted)	Items that include wild
^music	prefix-exact-match	Items that start with music
.mp3$	suffix-exact-match	Items that end with .mp3
!fire	inverse-exact-match	Items that do not include fire
!^music	inverse-prefix-exact-match	Items that do not start with music
!.mp3$	inverse-suffix-exact-match	Items that do not end with .mp3

All of these can be escaped by inserting a backslash \!, \^, \' and \$ (not mentioned in the README but tested in practice). The implementation here closely matches the behavior of both fzf and skim (which behave identical)

I found myself needing these recently. In particular, I found myself missing the inverse matches. I work on a circuit simulator (ngspice) which has some core code and then some device models (also just c code). When I was refactoring core parts of the simulator I wanted to use gr to find all references. However, this was polluted by thousands of references in the device models (all of which are inside the devices folder). So just being able to add ^devices in the picker was invaluable.

While looking into this I found myself warming up to the other options as well, for example with #5110 ^E can be used to filter only errors or !^H can be used to ignore hints.