Add large-run PerfettoSQL guidance to performance analysis docs

[Flamki]

Summary

Add practical guidance for analyzing large preCICE traces with PerfettoSQL and scripted Trace Processor queries.

What changed

• Updated content/docs/tooling/tooling-performance-analysis.md
• Added a new section: Analyze larger traces with PerfettoSQL
  • event inventory query
  • runtime aggregation for key preCICE events
  • expensive advance call query
  • per-track imbalance query
  • robust track_id to participant/process/thread mapping query
• Added a new section: Scripting trace analysis with Python
  • perfetto.trace_processor.TraceProcessor example for repeatable analysis
• Added installation hints for:
  • trace_processor_shell
  • Python package perfetto

Scope

• Content-only docs update
• No JS/CSS/layout/behavior changes

Context

• Relates to Add perfetto instructions for larger runs #631
• This PR focuses on large-run analysis guidance (PerfettoSQL + scripting).

Validation

• pre-commit run --files content/docs/tooling/tooling-performance-analysis.md (passed)
• docker run --rm -v "${PWD}:/srv/jekyll" -w /srv/jekyll jekyll/jekyll:4 bash -lc "bundle install && bundle exec jekyll build" (passed)