Simplify async doc examples

[clux]

There are two patterns we use in our docs at the moment:

    /// ```no_run
    /// use kube::{Api, Client};
    /// use k8s_openapi::api::core::v1::Pod;
    /// #[tokio::main]
    /// async fn main() -> Result<(), Box<dyn std::error::Error>> {
    ///     let client = Client::try_default().await?;
    ///     let pods: Api<Pod> = Api::namespaced(client, "apps");
    ///     let p: Pod = pods.get("blog").await?;
    ///     Ok(())
    /// }
    /// ```

e.g. kube-client/src/api/core_methods.rs

/// # async fn wrapper() -> Result<(), Box<dyn std::error::Error>> {
/// some_uncommented_async_operation().await?;
/// # Ok(())
/// # }

e.g. kube-runtime/src/events.rs

The first leads to the full async-setup visible in every example (via e.g. Api docs), and the second leads to business-action only (e.g. via EventRecorder docs)

We should align api docs so that we use the second pattern where possible. Comment out hidden imports, comment out the async wrappers. Deindent.

If we are documenting in kube_runtime we can also use more of the /// # let client: kube::Client = todo!(); pattern to focus on what the crate is about, not what kube_client is about.

To find places to replace search within the crates:

$ rg "tokio::main" kube-runtime
kube-runtime/src/controller/mod.rs
517:/// #[tokio::main]

kube-runtime/src/watcher.rs
692:/// #[tokio::main]
755:/// #[tokio::main]

affected crates:

• kube-runtime
• kube-client
• kube

(the ones in examples/e2e/readmes should be left in place)