feat(health checks): implement new kubernetes 1.2 health check features

[helgi]

Added the following new functionalities

• HEALTHCHECK_PERIOD_SECONDS - How often (in seconds) to perform the probe
• HEALTHCHECK_SUCCESS_THRESHOLD - How many probe runs need to be done after failure to consider it a success again
• HEALTHCHECK_FAILURE_THRESHOLD - How many times a probe should fail before considered being down

These defaults are all copied from kubernetes but may have to be adjusted due to our high default timeout, or we bring the default timeout down

Closes #251
Depends on #591 being merged first

[codecov-io]

Current coverage is 83.68%

Merging #600 into master will increase coverage by +0.29% as of 16c8511

@@            master    #600   diff @@
======================================
  Files           36      36       
  Stmts         2427    2433     +6
  Branches       351     351       
  Methods          0       0       
======================================
+ Hit           2024    2036    +12
+ Partial        116     114     -2
+ Missed         287     283     -4

Review entire Coverage Diff as of 16c8511

Powered by Codecov. Updated on successful CI builds.

[slack]

Can you drop a deis/workflow issue to track adding these to the docs?

[helgi]

Yeah, I have #570 but realised just now these are not ENV Vars :P I need to turn Health Checks into a section of the config instead of general configuration

[helgi]

Created deis/workflow#134 to track documentation needs

[helgi]

Wrote deis/workflow#135 for the documentation

[arschles]

This lgtm.

In case it helps anyone else, I asked @helgi for some additional information on what's going on here, so adding that info here in case it helps anyone else:

this PR adds the above env variables that are settable via config:set, which tell the controller how to configure app pods w.r.t. health check information. There is also a plan in #603 to move these config values to a top level config element, similar to what we do for tags.