Issue parsing kubernetes json

[lpic10]

Bug report

Using latest telegraf (built from master), kubernetes v1.2

Explanation: when a pod is stopped, the k8s API returns a null startTime for the podRef, as below:

{
			"podRef" : {
				"name" : "mypod-kxgfd",
				"namespace" : "default",
				"uid" : "2533197d-d110-11e6-b4b8-005056835c46"
			},
			"startTime" : null,
			"containers" : [{
					"name" : "mycontainer",
					"startTime" : "2017-01-02T17:24:14Z",
					"cpu" : {
						"time" : "2017-01-18T12:53:19Z",
						"usageNanoCores" : 0,
						"usageCoreNanoSeconds" : 0
					},
					"memory" : {
						"time" : "2017-01-18T12:53:19Z",
						"usageBytes" : 0,
						"workingSetBytes" : 0,
						"pageFaults" : 0,
						"majorPageFaults" : 0
					},
					"rootfs" : {
						"availableBytes" : 85094039552,
						"capacityBytes" : 107151884288
					},
					"logs" : {
						"availableBytes" : 2861916160,
						"capacityBytes" : 3103539200
					},
					"userDefinedMetrics" : null
				}
			]
		}

That causes the following error message in Telegraf:

2017-01-18T13:57:30Z E! ERROR in input [inputs.kubernetes]: Errors encountered: [Error parsing response: parsing time "null" as ""2006-01-02T15:04:05Z07:00"": cannot parse "null" as """]

That causes the whole kubernetes metrics to not be collected correctly if there is a single pod stopped.
The modification below has fixed it for me, but I'm not sure it is the best way of doing this:

--- a/plugins/inputs/kubernetes/kubernetes_metrics.go
+++ b/plugins/inputs/kubernetes/kubernetes_metrics.go
@@ -45,7 +45,7 @@ type CPUMetrics struct {
 // PodMetrics contains metric data on a given pod
 type PodMetrics struct {
        PodRef     PodReference       `json:"podRef"`
-       StartTime  time.Time          `json:"startTime"`
+       StartTime  *time.Time         `json:"startTime"`
        Containers []ContainerMetrics `json:"containers"`
        Network    NetworkMetrics     `json:"network"`
        Volumes    []VolumeMetrics    `json:"volume"`