Emulate cURL CLI flags: --retry N --retry-all-errors

Feature request for your consideration.

`curl` does not retry by default, but it can be instructed to do so with the command line flags `--retry` and `--retry-all-errors`, maybe `grpcurl` should do the same or something similar.

Background: I'm doing a `docker run --detach --name MyApp MyApp`

Followed by a `docker run --network container:MyApp fullstorydev/grpcurl 127.0.0.1:50051 MyService/MyRPC`, but I get something like `Failed to dial target host "127.0.0.1:50051": dial tcp 127.0.0.1:50051: connect: connection refused` because nor the container or the app can come up as fast as the next command/script/etc.

Rather than adding an extra step to "wait for MYAPP to be ready" I rather just slap a command line flag to `grpcurl`

disclaimer:
- I know is almost impossible to keep feature parity with the real `curl`
- I know this can be accomplished in other ways, I just think it would be a "nice to have" feature

Thanks!

The issue names no files or tests. Start by locating grpcurl's command-line flag handling and connection setup, then define how --retry and --retry-all-errors should handle the reported connection-refused startup case; done means the CLI can reliably retry that case with documented flag behavior.

Written by the indexing model from the issue text.