Fix bug in parsing domain from repository reference

[avtakkar]

regexp.DomainRegexp matches mixed-case and uppercase strings without . as well. For example, in /v2/SAMPLE/reponame/manifests/latest, the route handler would match SAMPLE as regexp.DomainRegexp and reponame as regexp.nameComponentRegexp. To parse the normalized name, an uppercase or mixed-case string should be considered a valid domain.

Fixes #2858
Redis cache enabled path invokes ParseNormalizedName which produces an error because rather than matching TESTNAME to the domain like the router does, it matches TESTNAME/reponame to the name component.

[GordonTheTurtle]

Please sign your commits following these rules:
https://github.com/moby/moby/blob/master/CONTRIBUTING.md#sign-your-work
The easiest way to do this is to amend the last commit:

$ git clone -b "avtakkar/fix-domain-split" git@github.com:avtakkar/distribution.git somewhere
$ cd somewhere
$ git commit --amend -s --no-edit
$ git push -f

Amending updates the existing PR. You DO NOT need to open a new one.

[codecov]

Codecov Report

Merging #2979 into master will decrease coverage by 0.04%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #2979      +/-   ##
==========================================
- Coverage   60.69%   60.64%   -0.05%     
==========================================
  Files         102      102              
  Lines        8044     8044              
==========================================
- Hits         4882     4878       -4     
- Misses       2511     2514       +3     
- Partials      651      652       +1

Flag	Coverage Δ
#linux	60.64% <100%> (-0.05%)	⬇️

Impacted Files	Coverage Δ
reference/normalize.go	82.02% <100%> (ø)	⬆️
registry/handlers/app.go	48.16% <0%> (-0.7%)	⬇️

---

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 1fb7fff...e89db8d. Read the comment docs.

[caervs]

Yeah. Another way to say it is when the first part of a ref is mixed case then you know that it must be a domain since repo names can't be mixed case.

LGTM

[dmcgowan]

Change looks good,

Can we get a test added to TestParseRepositoryInfo which explicitly matches the mixed case value as the domain?

[dmcgowan]

LGTM