feat: add Compare using go-cmp

[xiantank]

Add new matcher Compare, which use go-cmp to compare and we can pass cmp.Option as optional arguments.
here is example

Expect([]int{1, 2}).Should(Compare([]int{1, 2}))
Expect(myCustomType{s: "foo", n: 3, f: 2.0, arr: []string{"a", "b"}}).Should(Compare(myCustomType{s: "foo", n: 3, f: 2.0, arr: []string{"a", "b"}}, cmp.AllowUnexported(myCustomType{})))
Expect(myCustomType{s: "abc", n: 3, f: 2.0, arr: []string{"a", "b"}}).Should(Compare(myCustomType{s: "foo", n: 3, f: 2.0, arr: []string{"a", "b"}}, cmpopts.IgnoreUnexported(myCustomType{})))

related: #216, #264

[thediveo]

Just my tiny observation as I happened to look at go-cmp just these days (while working on checking file descriptor objects for equality). I remember that go-cmp thankfully documents its purposedly designed behavior of panicking, because it is intended to work without any matcher framework. It cannot rely on the usual error handling pattern that the Gomega matchers uses: matchers do not only return their "verdict" but also an error return value. I would, erm, expect that without handling panics the Compare matcher cannot be used in inverse expectations such as Expect(foo).NotTo(Compare()). If I didn't overlook something in your PR code (and please forgive me if I do and correct me please) then there's a inverse check using Equal but not one using Compare?

[xiantank]

Hi, I add some panic handling to make Compare matcher as usual error handling pattern, and remove a weird inverse check Equal (which I tried to show difference between Equal and Compare deal with different time location, but it's really weird to do so, thanks for pointing it out).
// please let me know if I misunderstood anything

[onsi]

hey @xiantank - can we rename it to BeComparableTo? That would flow better for the DSL Expect(foo).To(BeComparableTo(bar))

I think I'd, eventually, like to make it possible to configure Gomega so that Equal can be backed by go-cmp instead of reflect.DeepEqual. This is a great step in that direction, thanks!

[onsi]

thanks! looks good to me!

[onsi]

if you want to update the doc PR I can pull that in too