Fixes https://github.com/harfbuzz/harfbuzz/issues/1019

New numbers:

BENGALI: 353725 out of 354188 tests passed. 463 failed (0.130722%)
DEVANAGARI: 707261 out of 707394 tests passed. 133 failed (0.0188014%)
GUJARATI: 366353 out of 366457 tests passed. 104 failed (0.0283799%)
GURMUKHI: 60729 out of 60747 tests passed. 18 failed (0.0296311%)
KANNADA: 951300 out of 951913 tests passed. 613 failed (0.0643966%)
MALAYALAM: 1048136 out of 1048334 tests passed. 198 failed (0.0188871%)
ORIYA: 42327 out of 42329 tests passed. 2 failed (0.00472489%)
SINHALA: 271596 out of 271847 tests passed. 251 failed (0.0923313%)
TAMIL: 1091754 out of 1091754 tests passed. 0 failed (0%)
TELUGU: 970555 out of 970573 tests passed. 18 failed (0.00185457%)

Devanagari regressed because Uniscribe doesn't enforce the full set.

Tests added with the *-vowel-letters.txt files in tree and Noto fonts.

There should be a colon separating "(transfer full)" and the rest. Warned by g-ir-scanner.

Fixes https://github.com/harfbuzz/harfbuzz/issues/1102

* -Wshift-sign-overflow
* -Wmissing-prototypes

Fixes https://github.com/harfbuzz/harfbuzz/issues/1212

Fixes https://github.com/harfbuzz/harfbuzz/pull/933
https://github.com/harfbuzz/harfbuzz/issues/1012

Tibetan failures go from 0 to 2:

TIBETAN: 208467 out of 208469 tests passed. 2 failed (0.000959375%)

No machine change.

No behavior change.

Fixes https://github.com/harfbuzz/harfbuzz/issues/746

Fixes https://github.com/harfbuzz/harfbuzz/issues/1189

Fixes https://github.com/harfbuzz/harfbuzz/issues/1208

Missing "out" indicator. Affect gobject introspection.

The only Indic CM is U+0A75 GURMUKHI SIGN YAKASH, which Uniscribe doesn't treat
specially, and font designers design for that.  So, do the same.

Fixes https://github.com/harfbuzz/harfbuzz/issues/667

Based on experimenting with Uniscribe to extract grammar and categories.

Failures down from 44 to 35:

KHMER: 299089 out of 299124 tests passed. 35 failed (0.0117008%)

We still don't enforce the one-matra rule pre-decomposition, but enforce
an order and one-matra-per-position post-decomposition.

https://github.com/harfbuzz/harfbuzz/issues/667

Fixes https://github.com/harfbuzz/harfbuzz/issues/746

These are only relevant to Sinhala, because they decompose in other
cases.  The USE spec categorizes them all as VPst.  No idea why we
weren't following that before.