Batch Index Settings Update Requests

[original-brownbear]

Same as #79782 but for index settings. Each index setting upgrade does a reroute which can be very slow, we should have batching on these tasks to avoid queuing up lots of them and blocking CS updates for extended periods of time.

[elasticmachine]

Pinging @elastic/es-distributed (Team:Distributed)

[dakrone]

@original-brownbear in the interest of future scalability, should we make it a requirement that all submitted cluster state updates be batchable in the future? Perhaps deprecating the non-batched version of submission and building more helpers for batching in simple cases?