feat(en): expand Unicode symbol normalization

[HungerDeng]

Summary

Expand English text normalization for common Unicode math operators, currency symbols, micro-sign units, and general symbols, referencing the text replacement of the https://github.com/espeak-ng/espeak-ng/blob/master/dictsource/en_emoji

Changes

• Add mathematical operators and comparisons, including ≤, ≦, ≥, and ≧.
• Add currency symbols and micro-sign unit aliases under their semantic grammars.
• Add general arrows, typographic symbols, and music notation to the whitelist.
• Add shared English fixtures used by both pytest and Sparrowhawk normalization tests.

Testing

• PASS: python -m pytest -q tests/nemo_text_processing/en/test_math.py tests/nemo_text_processing/en/test_measure.py tests/nemo_text_processing/en/test_money.py tests/nemo_text_processing/en/test_whitelist.py --cpu --use_local_test_data --tn_cache_dir=/private/tmp/nemo_tn_cache (492 passed).
• Full explicit pytest collection found 1,874 tests but was interrupted after 10:34 during global grammar construction; no failures were reported before interruption.
• Sparrowhawk grammar export completed and produced the English FAR files. The Docker-backed Sparrowhawk test could not run because this host has no Docker daemon; the macOS Bash 3.2 script also reports unsupported ${VAR,,} syntax.

Notes for Reviewers

The distinct Unicode multiplication aliases (×, *, ·, ∗, ∙, and ⋅) intentionally remain mapped to times in math_operation.tsv; they are different input code points.