exposure analysis first branch

[shireenf-ibm]

issue : #236

[shireenf-ibm]

hi @adisos , in this new branch :

• an exposure-analysis flag is added (--exposure)
• new "fake" pods are added by the policy-engine , in this PR, scanning the policies for rules with only namespaceSelector and adding the relevant namespace and pod for each

• allowed conns will appear in the outputs when running with the flag (still need to do some enhancements for the outputs (in next PRs) , like:

  • in txt : writing the "exposure-analysis" lines in separate section/ at the end
  • in dot : we can see now the new pods in new namespaces (need to enhance the visibility and write the representative string we want)
  • in txt : we can see what labels the potential namespace should have (see example below)
  • wip: for unprotected pods, we see both conns with {any-namespace} and {namespace with potential labels} .. I need to eliminate conns with the second when the first appears (see example below)
  • some TODO's and comments are in the code (some are for next PRs)

  results example: