Support GPT-5.5 review workflow and clarify R table indexing

[johnzed0102]

Summary

• omit temperature for GPT-5/reasoning-model OpenAI structured review calls
• update LLM setup docs and examples to use gpt-5.5
• clarify R inspection docs around zero-based table_index vs paper table_number
• add table_index display to paper inspection helpers
• match variable-plausibility reviews by table_id to avoid sparse-review misalignment
• ignore local papers/ fixtures alongside existing outputs/ artifacts

Validation

• .venv/bin/python -m compileall table1_parser tests
• Rscript -e 'source("R/inspect_paper_outputs.R"); cat("ok\n")'

Notes

• pytest was not run because pytest is not installed in the local .venv.