Document community recommended community moderation policies

Following a discussion that has come up; it could be very useful if we were to have a "recommended" set of moderation policies that moderators in various communities could follow.

At a first pass, we can ask that the [Guidelines For Respectful Communication](https://haskell.foundation/guidelines-for-respectful-communication/) are used, and then we can recommend some moderation practices that we agree on.

To open the discussion; perhaps we could start with:

1. Moderators will actively enforce the guidelines,
2. We recommend reaching out to the poster and asking them to re-word/reconsider in line with the guidelines; in terms of acknowledging the other persons points; kindly expanding on disagreements, attempting to be curious about disagreements.
3. On occasion of repeated breaches, posts should be removed or hidden,
4. Flagging by community members should be encouraged,
5. There should be a regular process to raise community members into a pool of moderators; potentially also rotating moderator duties every year or two, to reduce burn-out,
6. Reminding users that moderation and respectful behaviour is a community effort,
7. Banning users should only be considered after repeated violations of the guidelines and an ability to reach an agreement in private conversation; all attempts should be made to avoid banning.

The wording of the above is not perfect; and probably missing many things; if you have suggestions please note them, and then we can attempt to draft something more complete!

### Todo

- [ ] Draft up something following the above
- [ ] Run it by a few Haskell communities/moderators and seek feedback/changes
- [ ] Discourse
- [ ] Reddit
- [ ] Some of the Discords?
- [ ] ???
- [ ] ???
- [ ] That's it!

Inizia con le Guidelines For Respectful Communication collegate e le pratiche di moderazione proposte in questa issue. Redigi una raccomandazione più completa, quindi raccogli feedback da Discourse, Reddit e dalle community Discord pertinenti, come indicato nel TODO. Il lavoro sarà considerato completato quando la bozza rifletterà il feedback della community e l'ambito rimanente sarà concordato.

Scritto dal modello di indicizzazione a partire dal testo della issue.