Audit the destructors of classes derived from net::URLRequestContextGetter

All classes that directly derive from net::URLRequestContextGetter should have "protected"
virtual destructors, not "public".

By having a public destructor, it becomes possible to stack
allocate the derived class. Stack allocating a RCTS is a
dangerous behaviour, since there may still be a caller who
holds a reference when the object goes out of scope.

BUG=none
TEST=none

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=131072

[Ryan Sleevi, 2012-04-04 00:42:00]

jam: content/
jamiewalch: remoting/
atwilson: chrome/browser/sync
mnissler: chrome/browser/policy
eroman: content/common/net

I noticed this after alexeypa committed http://crrev.com/130511 . If these
destructors were protected (as they are on the base class), it never would have
compiled.

Further, the lack of a virtual destructor for some of these is quite
questionable, so those are fixed as well.

This is the "short term" solution. The "long term" solution is figure out what
tooling changes (eg: clang) can be done to notice such a pattern and warn
appropriately.

The warnings would be:
- If you are-a RCTS, you should not have a public destructor.
- If you derive from a base class that has a virtual dtor, you have data
members, and you do not declare a virtual dtor (aka: -Wnon-virtual-dtor, which
I'm not sure why this wasn't caught)

[jam, 2012-04-04 01:02:24]

content lgtm

[Jamie, 2012-04-04 01:04:12]

remoting lgtm.

Regarding your virtual dtor remark, I agree that the derived classes should
specify virtual explicitly in keeping with the style guide, but I don't think it
will leak because the base class dtor is virtual.

[Andrew T Wilson (Slow), 2012-04-04 01:10:28]

browser/sync LGTM

[Mattias Nissler (ping if slow), 2012-04-04 09:16:09]

policy lgtm

[eroman, 2012-04-05 00:56:52]

lgtm

[commit-bot: I haz the power, 2012-04-05 00:59:37]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rsleevi@chromium.org/9958145/6001

[commit-bot: I haz the power, 2012-04-05 00:59:42]

Can't apply patch for file remoting/host/url_request_context.h.
While running patch -p1 --forward --force;
patching file remoting/host/url_request_context.h
Hunk #1 FAILED at 35.
1 out of 1 hunk FAILED -- saving rejects to file
remoting/host/url_request_context.h.rej

[commit-bot: I haz the power, 2012-04-06 00:25:48]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rsleevi@chromium.org/9958145/10001

[commit-bot: I haz the power, 2012-04-06 02:27:19]

Change committed as 131072