Right now the only workaround is to do some strange-ness within a pipeline where you create a pipeline and parallel the template along with a groovy script(In my case I included it within a global shared library).

So here are the steps I used:
1. Create a system log called "Kubernetes Log" with the `org.csanchez.jenkins.plugins.kubernetes` logger set to ALL
2. Create a pipeline job with a parallel statement which runs the template and loops through the following script until the template spins up:

import java.util.logging.Level
import java.util.logging.Logger
import hudson.logging.LogRecorderManager
import hudson.logging.LogRecorder
import java.util.logging.LogRecord
import hudson.util.RingBufferLogHandler;


def AgentName= searchString
List<LogRecord> records = new ArrayList<LogRecord>();

//Grabs the log manager
LogRecorderManager mgr = Jenkins.instance.getLog();

//Grabs the records
mgr.logRecorders.each{
   if (it.getValue().getName() == "KubernetesLog")
   {
     records = it.getValue().getLogRecords()
   }
 }

//Iterates over the record messages looking for the agent name
for (LogRecord r : records) {
  if (r.getMessage().contains(AgentName)){
    println(r.getMessage().toString())
  }
}

//Clears the logger
mgr.logRecorders.each{
   if (it.getValue().getName() == "KubernetesLog")
   {
     it.getValue().doClear()
   }
 }

This will show the messages relating to `searchString` which should be the name of your template.

We implemented a change to the kubernetes-plugin that will check the message of containers in the Waiting state. The message will contain the String "Back-off pulling image" when it can't locate the docker image (e.g. a badly defined docker image). We then grab the corresponding build job from the Jenkins Queue, print a message to the build's console output to notify users that they've specified a bad docker image, then cancel the build. Canceling the job and not simply labeling it as failed was necessary or else Jenkins would continuously re-try to create the Kubernetes pod using the bad docker image and fail. 
Our solution solves the problem of customers not knowing why their job is stuck in a perpetual limbo due to a bad docker image and not knowing / having permissions to view the kubernetes logs, as well as the problem of jobs being stuck in the aforementioned perpetual waiting state due to a malformed docker image. 

We are currently in the process of refining it and will submit a formal PR once that's ready. Any suggestions and comments would be appreciated.

UPDATE

Here is the PR with our suggested changes: https://github.com/jenkinsci/kubernetes-plugin/pull/440