Enable encoding detection for Polish text

Previously, some accented letters commonly used to write Polish text
were counted as 'rare' codepoints. Treat them as 'common' instead.

Thanks to Alec for pointing this out.

Author: alexdowad

ext/mbstring/common_codepoints.txt

@@ -3,6 +3,11 @@
 0x0020	0x007E # ASCII
 0x00A1	0x00AC # Pound sign, Yen sign, copyright sign...
 0x00AE	0x00FF # Accented Latin characters
+0x0104	0x0107 # Polish
+0x0118	0x0119 # Polish
+0x0141	0x0144 # Polish
+0x015A	0x015B # Polish
+0x0179	0x017C # Polish
 0x0300	0x030A # Diacritical marks
 0x0370	0x0377 # Greek
 0x037A	0x037F # Greek

ext/mbstring/rare_cp_bitvec.h

@@ -11,7 +11,7 @@
 
 static uint32_t rare_codepoint_bitvec[] = {
 0xffffd9ff, 0x00000000, 0x00000000, 0x80000000, 0xffffffff, 0x00002001, 0x00000000, 0x00000000,
-0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
+0xfcffff0f, 0xffffffff, 0xf3ffffe1, 0xe1ffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff,
 0xfffff800, 0xffffffff, 0xffffffff, 0x0300ffff, 0x0000280f, 0x00000004, 0x00000000, 0x00000000,
 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000,