wire/common: Implement binarySerializer based on sync.Pool

[kcalvinalvin]

sync.Pool implementation saves on the overhead of go channels and has the benefit of dynamically sizing the binaryFreeList.

sync.Pool is also used in the fmt package

[coveralls]

Pull Request Test Coverage Report for Build 530213972

• 174 of 179 (97.21%) changed or added relevant lines in 3 files are covered.
• 18 unchanged lines in 3 files lost coverage.
• Overall coverage increased (+0.01%) to 53.412%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
wire/common.go	146	151	96.69%

Files with Coverage Reduction	New Missed Lines	%
wire/common.go	2	97.27%
btcec/signature.go	3	92.48%
peer/peer.go	13	75.24%

Totals
Change from base Build 512434868:	0.01%
Covered Lines:	20869
Relevant Lines:	39072

---

💛 - Coveralls

[cfromknecht]

@kcalvinalvin this looks like it would be a nice companion to #1426. there will be some rebase conflicts but it'd be cool to see how much it improves the numbers in that PR.

[kcalvinalvin]

@kcalvinalvin this looks like it would be a nice companion to #1426. there will be some rebase conflicts but it'd be cool to see how much it improves the numbers in that PR.

I can rebase this PR on top of #1426. Would it be better to just make a PR to cfromknecht:wire-opts?

[BrannonKing]

A note on this: stack allocations would be faster than borrowing memory for all of the functions that return integers. There's no need to borrow anything on those.