End of experiment, we might get back to it later

Fixes https://github.com/harfbuzz/harfbuzz/issues/1159

Also, don't attach ZWNJ to previous cluster.  Closer to Unicode graphemes.

This adds a new grapheme bit.  Not used yet.

Part of https://github.com/harfbuzz/harfbuzz/issues/1159

As Khaled's suggestion, hard-coded font paths was only for my own testing.

Related #1219

Part of https://github.com/harfbuzz/harfbuzz/issues/1159

.

Add the default font path used by FreeBSD ports.

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.