Implemeting target assistants

- Workload's update_result stage has now been broken up into two parts:
  extract_results and update_output. This is to allow the assistant to
  pull output from the target in between the two stages.
- Updated assistant implementations for Linux and Android targets from
  the exisiting code.
- Extended target descriptor code to handle assistants and their
  parameters as well.
- Updated  the target manager to actually make use of the assistants.

wa/framework/execution.py

@@ -138,16 +138,21 @@ class ExecutionContext(object):
         self.current_job = self.job_queue.pop(0)
         self.current_job.output = init_job_output(self.run_output, self.current_job)
         self.update_job_state(self.current_job)
+        self.tm.start()
         return self.current_job
 
     def end_job(self):
         if not self.current_job:
             raise RuntimeError('No jobs in progress')
+        self.tm.stop()
         self.completed_jobs.append(self.current_job)
         self.update_job_state(self.current_job)
         self.output.write_result()
         self.current_job = None
 
+    def extract_results(self):
+        self.tm.extract_results(self)
+
     def move_failed(self, job):
         self.run_output.move_failed(job.output)