Moved `AWS_REGIONS` into `ec2` module
Created `ec2_connect` method in `ec2` module
Updated modules able to use `ec2_connect` and `AWS_REGIONS`

Line 276: no_reboot = dict(default=True, type="bool"),

This should really default to False (When AMIs are created the machine will reboot before copying the disk).

"When enabled, Amazon EC2 does not shut down the instance before creating the image. When this option is used, file system integrity on the created image cannot be guaranteed."

The `ec2_ami`, `ec2_elb`, `ec2_tag`, `ec2_vpc`, `route53`, and `s3` modules
all canonicalize the AWS access and secret key params as
`aws_access_key` and `aws_secret_key`. However, following the fixes for #4540,
those modules now use `get_ec2_creds` from `lib/ansible/module_utils/ec2.py`,
which requires access/secret key params to be canonicalized as
`ec2_access_key` and `ec2_secret_key`. As a result, AWS credentials passed
to those six modules as parameters are ignored (they instead always use
the AWS credentials specified via environment variables, or nothing).

So this change fixes those six modules to canonicalize the
AWS access and secret key params as `ec2_access_key` and `ec2_secret_key`,
allowing them to again accept AWS credentials passed via module params.

raise errors when omitted, updated changelog with new modules.

Calling `ec2.get_image` right after `ec2.create_image` may raise error
"InvalidAMIID.NotFound". This has happend roughly 1 time out of 10 for me.

Other people has bitten by this too:
- https://github.com/Netflix/aminator/commit/5707f100a07d3432027d1e7a3305567a2a7d2574
- http://stackoverflow.com/a/14794952
- https://bitbucket.org/utoolity/bamboo-aws-plugin/pull-request/22/baws-116-fix-ec2-image-task-failing-with/diff

No idea if I'm rebasing properly or not.  This is my first attempt.

Getting timeout errors this morning as AMIs are taking a bit longer to create, so I changed to timeout to 15 minutes