Remove support for Go 1.11 modules

[dylan-bourque]

Remove go.mod and add notes to README.md indicating opt-out for Go 1.11 Modules.

This PR probably also resolves #61, but not to the satisfaction of the original submitter. After much discussion, it's actually not supported to have a single import path for both module-aware and module-unaware consumers for v2+ packages.

[codecov-io]

Codecov Report

Merging #67 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master      #67   +/-   ##
=======================================
  Coverage   98.92%   98.92%           
=======================================
  Files           4        4           
  Lines         279      279           
=======================================
  Hits          276      276           
  Misses          2        2           
  Partials        1        1

---

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 9f278e1...2d35c48. Read the comment docs.

[theckman]

@dylan-bourque thank you for taking the time to do this. While doing some noodling on this today, while my relatives all spoke a language I don't understand around me, I realized that this is probably going to fuck over any library maintainer who has migrated to gofrs/uuid. (👋 @markbates) Anyone who tries to use their library will fail to resolve the /v3 import, and their only recourse will be to nag upstream or fork/patch.

I don't think it's a step we can take. I'm thinking we need something more akin to #66 in terms of code, with the README copy from this PR. However, one addition to that PR would be to add a blurb to the top of the README deprecating the /v3 import by a specific date. We should not include a go.mod file in /v3 either.

How about $now + 100 days? Gives everyone three months to migrate away from it, and a few days for us to merge without it going under 90 days. 😉

Edit: retained for transparency; we did not need to take this path and could just remove modules support. Everything should mostly be okay as a result, with new consumers not using the SIV (/v3) import.

[theckman]

cc: @itsjamie

[itsjamie]

That works for me. I'll reopen #66. @dylan-bourque it's off a branch in this repo, can you commit the README blurb there?

[dylan-bourque]

@itsjamie I thought the decision (in the #gofrs Slack channel) was to go with this PR (not creating the /v3 folder). I'm confused.

[itsjamie]

@dylan-bourque ignore me. I was confused :) 👍