Make etlas update robust

[rahulmutt]

As in #621, etlas update can crash if any steps of the update process tries to connect to a temporarily unavailable server. Instead, it should issue a warning that the given step did not succeed because the server was down and keep going to the remaining steps.

[jcolivo]

I'm submitting this feedback to let you know the output from my end.

I made the change in the config file -- changing send-metrics: False to send-metrics: True. I then ran etlas update, and got the following message in the prompt:

---

C:\Users\jc>etlas update
Downloading the latest package lists from:

• hackage.haskell.org
• etlas.typelead.com
  Updating binary package index.
  dieVerbatim: user error (etlas: Failed to download
  http://objects-us-west-1.dream.io/hackage-mirror/01-index.tar.gz : No Status
  Code could be parsed from response: An exception occurred during a WebClient
  request.
  )

---

I would also like to note that this message shows up even when send-metrics: is back to False.

[rahulmutt]

@jcolivo That problem looks related to: typelead/etlas#24. It seems to be a powershell related issue.

[jcolivo]

I provided a comment in: typelead/etlas#24. It's an issue related to installing curl via the Chocolatey Windows package manager, which was installed via the Windows PowerShell (Admin).