[Redis] Update endpoint returns a HTTP 504 Timeout

👋

We run nightly acceptance tests against Azure Public in various regions - a selection of these tests are spinning up a Redis Cache instance.

For one of these tests we first create a Redis Cache with RDB Backup Enabled via the API and then disable it by calling the Update endpoint which is returns a `HTTP 504` timeout. Unfortunately this test is flapping a lot, so it's hard to give an exact amount of time where this is broken, however looking at the logs I'd estimate that somewhere in the region of 30-40% of the calls to the Update endpoint return a `HTTP 504` from the server.

The Response we get back from the Go SDK is:

```
redis.GroupClient#Update: Failure responding to request: StatusCode=504 -- Original Error: autorest/azure: Service returned an error. Status=504 Code="GatewayTimeout" Message="The gateway did not receive a response from 'Microsoft.Cache' within the specified time period."
```

However that's merely wrapping the response from the API - which is returning a HTTP 504 - so from what I can tell this is an issue with the Redis API. Would it be possible for someone to take a look?

Thanks!

cc @TimLovellSmith @sarangan12

Start with the Redis Cache Update endpoint described in the report and review the Azure Public acceptance-test logs for the intermittent HTTP 504 responses. Compare calls that disable RDB Backup with successful updates and determine what evidence is needed to confirm the endpoint behaves reliably; the issue does not name a repository file or test to change.

Written by the indexing model from the issue text.