Add channel future timeout to be backwards compatible with older clients

[mitch2na]

[jenkinskurento]

Hi there, thanks for your Pull Request!

A Kurento member needs to verify that this patch is reasonable to test. In case it is, they should write a comment with the phrase test this please. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by Kurento members will still work. Regular contributors can be whitelisted to skip this step.

[JRuedas]

Hi, is there a plan to merge this fix?

[j1elo]

This is full of non relevant changes that introduces noise and hides the actual, meaningful changes. Please clean up and don't introduce stylistic or whitespace-only changes. Please mind the contributing guidelines.

Apart from that:

• What problem is this trying to solve?
• What are the consequences of the changes?

[JRuedas]

This is full of non relevant changes that introduces noise and hides the actual, meaningful changes. Please clean up and don't introduce stylistic or whitespace-only changes. Please mind the contributing guidelines.

Apart from that:

• What problem is this trying to solve?
• What are the consequences of the changes?

Hi, this PR solves an important problem when you are using KMS in a production environment. The problem is that the Kurento Client cannot reconnect to the KMS after this one has stopped and started again.

When the KMS stops working, the Kurento Client tries to start a new Netty channel and gets stuck there, so even if the KMS is working again, it will never connect again.

[Forcada]

We are working with the changes of #12 because they fixed KurentoClient hanging when the connection with the KMS drops.

This pull request seems to work in the same way for the reconnection with the KMS too.