Add Celtic Sea worked example to Getting started guide (#450)

[gustavdelius]

Closes #450.

What

The "Getting started" guide routed readers to external course material and cheat sheets for a real, end-to-end example. This adds one self-contained worked example on the page itself — a new "A worked example: the Celtic Sea" section that runs the whole mizer workflow against a real ecosystem:

1. Build a multi-species model from species parameters, an interaction matrix and gear definitions (newMultispeciesParams()).
2. Find the steady state (steady(), plotSpectra()).
3. Calibrate to observed biomass (calibrateBiomass() → alternating matchBiomasses()/matchGrowth()/steady()), verified with plotBiomassObservedVsModel().
4. Check against observed yield as an independent test (plotYieldObservedVsModel()).
5. Set resilience to fishing (setBevertonHolt()).
6. Project a fishing scenario — an effort sweep — and interpret an indicator: the community sustainable-yield curve, which shows the fishery sits slightly past the effort that maximises total long-term catch.

The model is the Celtic Sea model from the Build part of the course, based on Spence et al. (2021); the small input data files are downloaded from the course repository at build time.

Also addresses the issue's smaller point: the guide's first newMultispeciesParams() example no longer prints the h/von-Bertalanffy default-parameter notes, which were alarming to new users out of context. They are now explained calmly in prose instead.

Notes

• mizer.Rmd is .Rbuildignored (only mizer_vignette.Rmd ships to CRAN), so this section is built only for the website by pkgdown and does not run during R CMD check — the build-time download and ~1–2 min calibration are fine there.
• Added the Spence et al. (2021) reference to reflib.bib.
• Verified the full guide knits to HTML end-to-end: the calibration and scenario code produce real output and plots, the citation resolves, and the in-page anchor links to the new section resolve.

🤖 Generated with Claude Code

https://claude.ai/code/session_01T98LXny6oRNBMy9kHubpo4