The newly introduced normalizer in the funders vocab generates two tokens when acronyms contain diacritics

[wgresshoff]

Package version (if known):

Since v5.1.0

Describe the bug

The normalization for new or updated entries in the funders index leads to the following problem:

The normalization token stream is expected to produce exactly 1 token, but got 2+ for analyzer analyzer name[accent_normalizer], analyzer [org.opensearch.index.analysis.CustomAnalyzer@33ac3672], analysisMode [ALL] and input "FüAkBw"

The document will not be indexed.

Steps to Reproduce

1. Run invenio index destroy --yes-i-know
2. Run invenio index init
3. Run invenio rdm rebuild-all-indices

Expected behavior

All records and vocabularies should be reindexed.