Add support for configurable timeouts in AWS OpsWorks Instances.

[charles-at-geospock]

We have found that the creation and update of the OpsWorks Instances often
times out with the default 10 minute timeout. This change introduces a
configuration for 'Create', 'Delete' and 'Update' timeouts. The actual
implementation is required to start or stop the instance, which is the
operation which times out, and which of these is performed (particularly
for the 'Update' operation) is dependant on the change needed.

The change introduces the timeouts, adds a simple test that the format
is parsed, and provides documentation for these new parameters.

[stack72]

Thanks so much for moving this PR and for the functionality update :) This LGTM!

% make testacc TEST=./aws TESTARGS='-run=TestAccAWSOpsworksInstance_'
==> Checking that code complies with gofmt requirements...
TF_ACC=1 go test ./aws -v -run=TestAccAWSOpsworksInstance_ -timeout 120m
=== RUN   TestAccAWSOpsworksInstance_importBasic
--- PASS: TestAccAWSOpsworksInstance_importBasic (139.33s)
=== RUN   TestAccAWSOpsworksInstance_UpdateHostNameForceNew
--- PASS: TestAccAWSOpsworksInstance_UpdateHostNameForceNew (168.26s)
PASS
ok  	github.com/terraform-providers/terraform-provider-aws/aws	307.625s

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thanks!