Add support for retry logic when timestamp errors are encountered

[japarson]

From https://github.com/Azure/azure-code-signing-action/issues/23:

We are using this action to sign our binaries. We are signing lots (~120) binaries at a time. On a fairly regular basis, we get errors.

Some errors look like this (but for seemingly arbitrary files):

SignTool Error: The specified timestamp server either could not be reached or returned an invalid response.

SignTool Error: An error occurred while attempting to sign: D:\a\

We should add retry logic to the underlying PowerShell Module for when we encounter sporadic timestamping errors. Looking at other tools, the convention appears to be to have 2 separate parameters:

• The number of retries for timestamping (default 3)
• The number of seconds to wait between timestamping retries (default 10s)

I'm trying a different time server http://timestamp.digicert.com/, which stackoverflow.com/questions/69016131/code-signing-which-timestamp-servers-are-more-reliable says is a reliable one. This is also the server we were using to sign our binaries before moving to azure code signing.

[aeisenberg]

Our internal workflow is now using the default time server again and things are working fine for us. As long as this is the case, we don't need this enhancement any more.