Be equally forgiving of leading & trailing whitespace in number parsing

[Tetralux]

Follow-up from #20588.

Before #20588, the parsing code for integers was forgiving of leading and trailing whitespace.
The boolean parsing code was not.

After #20588, the integer parsing code was still forgiving of whitespace.
The boolean parsing code was still not.

This PR makes all three forgiving of all leading and trailing whitespace, as determined by isspace.

cc @tkelman

[tkelman]

over indented by 1

[tkelman]

@nanosoldier runbenchmarks(ALL, vs = ":master")

[Tetralux]

@tkelman Oops. Might have inadvertently cancelled Nanosoldier by pushing?

[KristofferC]

Nope. It will post when done.

[nanosoldier]

Your benchmark job has completed - possible performance regressions were detected. A full report can be found here. cc @jrevels

[Tetralux]

Hmmm...

["array","index", ("sumvector_view","SubArray{Float32,2,Array{Float32,3},Tuple{Int64,Base.Slice{Base.OneTo{Int64}},Base.Slice{Base.OneTo{Int64}}},true}")]

I'm not sure I understand what this means. Something about a slice? :/
Can one of you point me to where I can see this benchmark?

[StefanKarpinski]

That seems unrelated since we're definitely not doing any parsing during view indexing.

[Tetralux]

@StefanKarpinski That's what was confusing me.
RTM?

[Tetralux]

Thank you. 😄