Docs: clarify installation notes and GNN vs CNN expectations

[CodeVishal-17]

Description

This PR adds a small documentation section clarifying common installation pitfalls
and setting expectations around when GNN-based models are most useful compared to CNNs.

Motivated by newcomer experience and discussion in #167.
No code or behavior changes are introduced.

Fixes #

How Has This Been Tested?

N/A – documentation-only changes.

• Yes

If your changes affect data processing, have you plotted any changes? i.e. have you done a quick sanity check?

• Yes

Checklist:

• My code follows OCF's coding style guidelines
• I have performed a self-review of my own changes
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works (not applicable)
• I have checked my changes and corrected any misspellings

[CodeVishal-17]

I checked the pre-commit failure in detail.

The failing ruff checks are reporting pre-existing docstring
violations (D100–D107) in files unrelated to this PR
(e.g. graph_weather/models/aurora/*), which were not modified here.

This PR is documentation-only, so I’ve avoided touching unrelated
Python files. Happy to adjust if there’s a preferred way to handle
pre-commit for docs-only PRs.