resource/aws_autoscaling_policy: Properly read step_adjustment into Terraform state

[bflad]

This pull request builds on top of #7195 and only requires review of 5fd80c1

Previously, the resource was silently failing to read this attribute and therefore unable to perform drift detection or properly import the resource (currently being added).

--- FAIL: TestAccAWSAutoscalingPolicy_basic (85.72s)
    testing.go:538: Step 2 error: ImportStateVerify attributes not equivalent. Difference is shown below. Top is actual, bottom is expected.

        (map[string]string) {
        }

        (map[string]string) (len=4) {
         (string) (len=17) "step_adjustment.#": (string) (len=1) "1",
         (string) (len=54) "step_adjustment.2042107634.metric_interval_lower_bound": (string) (len=1) "2",
         (string) (len=54) "step_adjustment.2042107634.metric_interval_upper_bound": (string) "",
         (string) (len=45) "step_adjustment.2042107634.scaling_adjustment": (string) (len=1) "1"
        }

Returning the error from d.Set():

--- FAIL: TestAccAWSAutoscalingPolicy_basic (14.03s)
    testing.go:538: Step 0 error: Error applying: 1 error occurred:
        	* aws_autoscaling_policy.foobar_step: 1 error occurred:
        	* aws_autoscaling_policy.foobar_step: error setting step_adjustment: step_adjustment.0.metric_interval_lower_bound: '' expected type 'string', got unconvertible type 'float64'

Output from acceptance testing after adjustments:

--- PASS: TestAccAWSAutoscalingPolicy_zerovalue (44.61s)
--- PASS: TestAccAWSAutoscalingPolicy_SimpleScalingStepAdjustment (46.60s)
--- PASS: TestAccAWSAutoscalingPolicy_TargetTrack_Predefined (48.75s)
--- PASS: TestAccAWSAutoscalingPolicy_disappears (72.73s)
--- PASS: TestAccAWSAutoscalingPolicy_TargetTrack_Custom (74.47s)
--- PASS: TestAccAWSAutoscalingPolicy_upgrade (76.43s)
--- PASS: TestAccAWSAutoscalingPolicy_basic (82.25s)

[mbfrahry]

LGTM after travis passes

[bflad]

Ha, we even had errant unit testing in this case:

--- FAIL: TestFlattenStepAdjustments (0.00s)
    structure_test.go:820: expected metric_interval_lower_bound to be 1.0, but got %!d(string=1)

I'll correct the unit testing. 😄

[bflad]

This has been released in version 1.57.0 of the AWS provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading.

[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!