CLDR-17865 Fix some bad currencyFormats/unitPattern items

[pedberg-icu]

CLDR-17865

• This PR completes the ticket.

Fixed most of the cases mentioned in CLDR-17865 where there are inconsistencies among the currencyFormats/unitPattern items for different plural cases in a locale. Note that the pattern in root for other and the patterns in English for other and one are all "{0} {1}" which produces results like "1 Euro" and "3 Euros". Many of the discrepancies in the ticket were for cases that had "{0} {1}" for the non-other cases but "{1} {0}" for the other case. The TC discussion was to prefer the other case pattern which would produce results whose English equivalent would be something like "Euros 3" (vs "1 Euro" which might be the desired result for one in some locales). I checked the Survey Tool votes and discussion (no discussion for any of these cases) to decide what to do.

• blo (Anii) had [(Zero, "{0} {1}"), (One, "{0} {1}"), (Other, "{1} {0}")] but the only vetter votes were for the other entry, so I used that for all cases.
• ceb (Cebuano) had [(One, "{0} {1}"), (Other, "{1} {0}")] but the only vetter (from Google) voted explicitly for each of those patterns, so I did not change the data.
• es_GT (Spanish, Guatemala) overrode the inherited value for other, replaced with up arrows.
• si (Sinhala) had [(One, "{0} {1}"), (Other, "{1}{0}")]; the one pattern only had votes from a Google vetter but the other pattern had votes from both Google and Microsoft, so used that for the one case as well.
• sw (Swahili) had [(One, "{0} {1}"), (Other, "{1} {0}")] but the only vetter (from Microsoft) voted explicitly for each of those patterns, so I did not change the data.
• vec (Venetian) had [(One, "{0} {1}"), (Other, "{0}\u{202f}{1}")] with NNBSP in the other case; the Venetian org voted for each of the cases, but I used the other pattern for one as well (same ordering, just the addition of NNBSP).

Filed CLDR-17873 to add a test for this.

ALLOW_MANY_COMMITS=true

[macchiati]

I've been thinking that, like how we have the hidden characters shown below the items, we could use placeholder substitution, so {1} {0} would turn into {NUMBER} {CURRENCY_NAME} or something like that.

[AEApple]

I've been thinking that, like how we have the hidden characters shown below the items, we could use placeholder substitution, so {1} {0} would turn into {NUMBER} {CURRENCY_NAME} or something like that.

We really should, but we need to be careful with placeholder naming. I have seen a lot of localization quality problems when the placeholders are named in a way that leads a language specialist to use the wrong preposition based on their interpretation of the placeholder name etc.