This PR adds extension support and implements the ICU extension. The ICU extension uses the minimal ICU collation library to add support for collations of various locales, rather than just the simple NOCASE/NOACCENT collations that are currently supported.

The extension is optional, and is by default not build. It can be build by passing the flag -DBUILD_ICU_EXTENSION=1 to CMake. The extension can be loaded into a database as follows:

DuckDB db;
db.LoadExtension<ICUExtension>();

Note that for now only statically linked extensions are supported, i.e. the extension must be compiled into the program. The current extension API does not (yet?) support loading extensions from shared libraries/DLLs.

Loading the extension allows a number of new collations to be used. The total list can be viewed using the PRAGMA collations command or queried using the pragma_collations function.

These are the supported collations:

// [af, am, ar, as, az, be, bg, bn, bo, bs, bs, ca, ceb, chr, cs, cy, da, de, de_AT, dsb, dz, ee, el, en, en_US, en_US, eo, es, et, fa, fa_AF, fi, fil, fo, fr, fr_CA, ga, gl, gu, ha, haw, he, he_IL, hi, hr, hsb, hu, hy, id, id_ID, ig, is, it, ja, ka, kk, kl, km, kn, ko, kok, ku, ky, lb, lkt, ln, lo, lt, lv, mk, ml, mn, mr, ms, mt, my, nb, nb_NO, ne, nl, nn, om, or, pa, pa, pa_IN, pl, ps, pt, ro, ru, se, si, sk, sl, smn, sq, sr, sr, sr_BA, sr_ME, sr_RS, sr, sr_BA, sr_RS, sv, sw, ta, te, th, tk, to, tr, ug, uk, ur, uz, vi, wae, wo, xh, yi, yo, zh, zh, zh_CN, zh_SG, zh, zh_HK, zh_MO, zh_TW, zu]

These correspond to the collation rules found here.