First bite of implementing span

[gongminmin]

Description

Implements span by following P0122R7 and P1024R3 Enhancing span Usability.

Related work item: #4

Checklist:

• I understand README.md.
• If this is a feature addition, that feature has been voted into the C++
  Working Draft.
• Any code files edited have been processed by clang-format 8.0.1.
  (The version is important because clang-format's behavior sometimes changes.)
• Identifiers in any product code changes are properly _Ugly as per
  https://eel.is/c++draft/lex.name#3.1 .
• Identifiers in test code changes are not _Ugly.
• Test code includes the correct headers as per the Standard, not just
  what happens to compile.
• The STL builds and test harnesses have passed (must be manually verified
  by an STL maintainer before CI is online, leave this unchecked for initial
  submission).
• This change introduces no known ABI breaks (adding members, renaming
  members, adding virtual functions, changing whether a type is an aggregate or
  trivially copyable, etc.). If unsure, leave this box unchecked and ask a
  maintainer for help.

[StephanTLavavej]

Thank you! As we are working on getting our tests and CI active, our full review will be delayed until then.

[miscco]

Given the discussion in #104 wouldnt span also need the special non-member overloads for begin and end wrt to rvalues

[CaseyCarter]

Given the discussion in #104 wouldnt span also need the special non-member overloads for begin and end wrt to rvalues

Yes, there have been substantial changes to span in the working draft in addition to P1024R3 since P122R7 was merged.

This header appears to be derived from the MIT-licensed GSL implementation of span.

[BillyONeal]

This header appears to be derived from the MIT-licensed GSL implementation of span.

I don't think we can accept this change in any form.

[StephanTLavavej]

Filed #117 as we haven’t described our license concerns yet (we didn’t expect to receive major PRs this early).

[StephanTLavavej]

Closing this PR; due to license concerns and Working Paper churn, we expect to implement span from scratch in the near future, instead of basing it on GSL's implementation in any way.