docs: add admin documentation (#436)

Fixes #422 

I tried to keep it short and concise. Please let me know if anything
important is missing and I can add it.
I think the documentation in the [Snakemake Cluster
Profiles](https://github.com/snakemake/snakemake-cluster-profiles) repo
will improve in the future and will also contain valuable information
for HPC admins.


<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **Documentation**
* Added a short guide for HPC cluster administrators covering global vs.
cluster-specific Snakemake profiles, the two-part profile structure
(partition config and executor-oriented cluster profile), where to find
template profiles, and user installation options (Conda, Spack,
Easybuild). No changes to runtime behavior or plugin error handling.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Co-authored-by: Christian Meesters <meesters@uni-mainz.de>
Co-authored-by: Christian Meesters <cmeesters@users.noreply.github.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>

Author: 3 people

docs/further.md

@@ -836,3 +836,18 @@ $ salloc -A <your SLURM account> -p <your partition> \
 You need to consider the extra waiting time for jobs to start on your cluster, when choosing a value for the `--time` flag.
 
 Note that `salloc` opens a subshell. Exit it before starting another interactive job in the same terminal -- even if the previous job has completed.
+
+### Short Guide for HPC Cluster Administrators
+
+As an HPC cluster administrator, you can help your users to run Snakemake seamlessly on your cluster.
+Consider providing a global Snakemake profile and storing it at `/etc/xdg/snakemake` on login and cluster nodes. 
+Such a configuration will not interfere with other workflow management systems and can always be overwritten by a user - it merely defines sensible defaults.
+
+Cluster-specific profiles consist of two files:
+- the Snakemake-specific partition configuration
+- and a cluster profile which contains information about this SLURM executor, potential file system latency to consider, mount points for temporary files (e.g. a node-local scratch) and more.
+
+To obtain a template for a cluster-specific profile, search at the [Snakemake cluster profiles repository](https://github.com/snakemake/snakemake-cluster-profiles). This repository not only provides configuration templates, but also maintainer contacts and deployment hints.
+
+Your users will install Snakemake by Conda, add this Slurm executor plugin, and source code for their workflows. Alternatively, Snakemake is provided by [Spack](https://packages.spack.io/package.html?name=snakemake) and [Easybuild](https://docs.easybuild.io/version-specific/supported-software/s/snakemake/), which both update regularly.
+