Ignore traces

This adds the ability to ignore traces to the backend. 
The core logic is in golden/go/types/ignorestore.go
The other code changes are refactors. 

BUG=skia:

Committed: https://skia.googlesource.com/buildbot/+/8e3738bfbfb38c4ffff87b063569e096f5b80faf

[stephana, 2015-01-30 16:58:14]

stephana@google.com changed reviewers:
+ jcgregorio@google.com

[jcgregorio, 2015-01-30 17:37:28]

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
File golden/go/analysis/analysis.go (right):

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:44: IgnoreRules []*types.IgnoreRule
This doesn't look like it's used anywhere.

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:429: // for _, r := range rules {
commented out ?

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:448: // DeleteIgnoreRule delets the ignore rule
and recalculates the state of the
deletes

https://codereview.chromium.org/884943003/diff/100001/golden/go/types/ignores...
File golden/go/types/ignorestore.go (right):

https://codereview.chromium.org/884943003/diff/100001/golden/go/types/ignores...
golden/go/types/ignorestore.go:85: // BuildRuleMatcher, see IgnoreStore
interface.
RuleMatcher seems excessively complex. Why not:

  type RuleMatcher func(tr types.Trace) bool

[stephana, 2015-01-30 20:27:45]

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
File golden/go/analysis/analysis.go (right):

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:44: IgnoreRules []*types.IgnoreRule
On 2015/01/30 17:37:28, jcgregorio wrote:
> This doesn't look like it's used anywhere.

It's used below by addIgnoreRules. It's not used in the output yet.
See TODO I added in ListIgnoreRules(...)

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:429: // for _, r := range rules {
On 2015/01/30 17:37:28, jcgregorio wrote:
> commented out ?

Added TODO.

https://codereview.chromium.org/884943003/diff/100001/golden/go/analysis/anal...
golden/go/analysis/analysis.go:448: // DeleteIgnoreRule delets the ignore rule
and recalculates the state of the
On 2015/01/30 17:37:28, jcgregorio wrote:
> deletes

Done.

https://codereview.chromium.org/884943003/diff/100001/golden/go/types/ignores...
File golden/go/types/ignorestore.go (right):

https://codereview.chromium.org/884943003/diff/100001/golden/go/types/ignores...
golden/go/types/ignorestore.go:85: // BuildRuleMatcher, see IgnoreStore
interface.
On 2015/01/30 17:37:28, jcgregorio wrote:
> RuleMatcher seems excessively complex. Why not:
> 
>   type RuleMatcher func(tr types.Trace) bool

I choose map[string]string instead of trace as an argument because it makes
testing easier (I don't need to set up a trace to test).
 
It's possible that a trace matches multiple rules and I want to be able to
provide statistics for each rule (see ListIgnoreRules function), that's why I
need to return the matched rules. 
The boolean is a convenience features so I don't have to check the length of the
returned array.

[jcgregorio, 2015-01-30 20:46:14]

lgtm

[stephana, 2015-01-30 20:47:19]

The CQ bit was checked by stephana@google.com

[commit-bot: I haz the power, 2015-01-30 20:47:40]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/884943003/140001

[commit-bot: I haz the power, 2015-01-30 20:47:48]

Committed patchset #8 (id:140001) as
https://skia.googlesource.com/buildbot/+/8e3738bfbfb38c4ffff87b063569e096f5b8...