Valid number string predicates

[kusamakura]

Adds the following string predicates: ValidInt, ValidLong, ValidDouble, ValidBigInt, and ValidBigDecimal.

I initially implemented this with a typeclass-based approach, e.g. ParsableString[N], which the Validate instance would use, but it seemed to be double the work than just creating new predicates and the corresponding Validate instances. Both seem extensible enough, though I surmise that a typeclass-based approach would be a little less verbose, since supplying the typeclass instances would be enough to get started. @fthomas, if you think that that's a more sensible approach, then I can always update this PR.

I'm also thinking of adding inference instances for these, but I'll hold off for now until I can find a good way of doing it without copious amounts of copy and paste.

[codecov]

Codecov Report

Merging #379 into master will increase coverage by 0.23%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #379      +/-   ##
==========================================
+ Coverage   96.38%   96.62%   +0.23%     
==========================================
  Files          41       41              
  Lines         498      503       +5     
  Branches       11       10       -1     
==========================================
+ Hits          480      486       +6     
+ Misses         18       17       -1

Impacted Files	Coverage Δ
...red/src/main/scala/eu/timepit/refined/string.scala	100% <100%> (ø)	⬆️
...c/main/scala/eu/timepit/refined/cats/package.scala	100% <0%> (+33.33%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 41e8946...40c0cc9. Read the comment docs.

[fthomas]

Thanks @kusamakura, this looks great!

The ParseableString[N] approach is an interesting idea. Instead of defining a predicate and the Validate instance you just need to define the ParseableString instance. But as you already noted, the latter is just a little bit less work and probable a little bit harder to understand.