Allow to use `crons` / `consumers` with raw GNU/Linux

[nfroidure]

Issue

Currently, creating crons / consumers can be done only for GCP/AWS targets. That said, in order to deploy Whook projects on simple GNU/Linux machines, we could create the same kind of mechanism to declare crons / consumers and then have commands that create the crontabs / consumers required (cron, systemd may help).

Technical tasks

Create a @whook/gnu-linux plugin that:

• [ ] provides a wrapper for crons and a command to install it locally (and eventually replay it)
• [ ] provides wrappers for consumers (kafka, mqtt, rabbitmq...), commands to install it and to test it locally (probably via systemd, see https://opensource.com/article/20/7/systemd-timershttps://opensource.com/article/20/7/systemd-timers )
• [ ] provides something (a command ?) that install the HTTP server via systemd
• [ ] provides something that manage logs rotates (cron + logrotate / systemd timer)