Posterior prediction, forecasting, and output handling tutorial

We have some of these the above mixed in throughout the tutorials (e.g. https://cdcgov.github.io/PyRenew/tutorials/building_multisignal_models.html#running-the-model) but it would be good to have a dedicated tutorial with attention to

• Performing posterior prediction (especially for forecasting)
• Handling time and other axis dimensions when extracting posterior draws
• Using arviz.extract() (and maybe polarbayes)?

Inspired by feedback from @confunguido and @KOVALW. Tagging @damonbayer and @cdc-mitzimorris for awareness.

Start with tutorials/building_multisignal_models.html#running-the-model and review how the existing tutorials handle model output. Create a dedicated tutorial covering posterior prediction and forecasting, time and other axis dimensions when extracting draws, and arviz.extract() with polarbayes if appropriate. Done means the tutorial explains these workflows clearly with runnable examples.