Data-driven Clusters v4.1 page (11 clusters from Google Sheet) (#720)

* Add data-driven Clusters v4.1 page with 11 clusters from Google Sheet

Replace the 7 hardcoded cluster markdown files with a data-driven approach
that reads from a generated JSON file (clusters_v4.json). The data originates
from the FORRT Clusters v4.1 Google Doc and is parsed into a Google Sheet,
then exported as JSON for Hugo to consume at build time.

Key changes:
- New script (parse_clusters_to_sheet.py) that parses the GDoc, resolves
  DOIs via doi.org for clean APA references + BibTeX, writes to Google Sheet,
  and exports JSON for Hugo
- New Hugo shortcode (clusters_display.html) renders all clusters with
  sidebar navigation, tabbed sub-clusters, and full-text search
- Updated intro text to reflect 11 clusters (was 9)
- Deactivated old cluster1-7.md files (replaced by data-driven rendering)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Disable site-wide search (replaced by clusters page search)

The clusters page now has its own full-text search that covers
clusters, sub-clusters, and all references. The site-wide Academic
search is redundant and has been disabled.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Revert "Disable site-wide search (replaced by clusters page search)"

This reverts commit 9ad47b9.

* enhancement:clusters page

* rm: left custom search, enhancement of the UI

* jump to active when tab clicked of sub cluster

* SEO booster for clusters

* Simplify clusters intro: concise text + lightbox image hero

Replace verbose multi-paragraph intro with a compact two-column layout
(text left, clickable thumbnail right) and remove syllabus section.
Adds lightbox overlay with magnifying glass hint for discoverability.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Update clusters intro: 100+ scholars, remove stale date

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* UI polish: 3-col sub-cluster grid, full-width layout, update badge

- Sub-cluster tabs in fixed 3-column grid with fixed height and centered text
- Remove reference counts from tabs
- Full-width layout for clusters display section
- Darker background on inactive tabs
- "Update pending" badge on clusters diagram

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* Clean up clusters code: remove dead CSS/JS, fix redundancies, align nav behavior

- Remove ~145 lines of dead .clusters-cluster-subpage CSS and ~65 lines
  of dead JS (element never exists in DOM)
- Remove unused list.html template and .clusters-title-toolbar CSS
- Extract duplicated color arrays to shared partial (colors.html)
- Remove duplicate intro text from _index.md (was inconsistent with intro.html)
- Fix sub-cluster nav on individual pages to use in-page anchors for the
  active cluster, aligning scroll behavior with the hub page
- Fix redundant .toLowerCase() in search
- Add min-height on .cluster-tab-content to keep footer below fold
  when viewing short sub-clusters

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Richard Dushime <45734838+richarddushime@users.noreply.github.com>
Co-authored-by: richarddushime <mudaherarich@gmail.com>

Author: 3 people

.gitignore

@@ -271,3 +271,6 @@ scripts/forrt_contribs/tenzing_failures.json
 
 # Bibtex to APA converter output
 bibtex_to_apa/node_modules/
+
+# DOI lookup cache for cluster parsing script
+scripts/doi_cache.json