* Support semicolon as extension character while parsing

* Add notes to pending_code_changes.txt

* JS port: Support semicolon as extension character while parsing

* Update comments in phonenumberutil.js

This excludes editing the proto file itself, or the auto-generated or hand-crafted fields that are derived from it.

Metadata changes:
-- Drops the "-1" that was erroneously included in possible lengths before (didn't break anything, but was wrong) - it was a possibleLength of a sub-component so got added to the generalDesc possibleLengths
-- possibleNumberPattern no longer inherited: we don't use this anyway, we will do another CL soon to stop including it at all in the generated metadata
-- exampleNumber is no longer set on fixed-line and mobile elements from the generalDesc

XML file changes:
-- Stopped specifying "NA" and "-1" specifically for fixed-line and mobile blocks; now they are treated as every other type of phone number: if missing, don't fill them in from generalDesc, but leave them missing.

Code changes:
-- Stop using the exampleNumber on generalDesc for non-geo entities, but look at their phonenumber descs - the exampleNumber won't be stored on the generalDesc anymore. This affects porters if they either copied our build logic or used our built metadata in some way; they should update this method in their port too.

Changing PhoneNumberUtil to use the possibleLengths information, not the reg-exes.

Note the API is not changing, but the metadata is now somewhat stricter for
many countries, since before we applied only a minimum and maximum length for
most countries, and now we specify exactly which lengths are possible.

This has a flow-on effect when parsing, since we decide whether to do certain
operations like strip a national prefix based on whether the number is a
possible length before/after - when parsing, if the number is shorter than the *national* pattern, we no longer strip the national prefix.

Affected countries:
AD (7 digits now invalid)
AM (7 digits now invalid)
AR (9 digits now invalid)
AZ (8 digits now invalid)
BG (4 digits now valid for local-only numbers)
BJ (5-7 digits now invalid)
CC/CX (5 digit numbers now possible: this should always have been the case, but the generalDesc was wrong and didn't reflect its child elements. We now calculate it based on them, which allows 5 digit numbers.)
CO (9 digits now invalid)
CR (9 digits now invalid)
ET (8 digits now invalid)
GE (7 and 8 digits now invalid)
GH (8 digits now invalid)
IL (5 and 6 digits now invalid)
IM/JE/GG (7, 8 and 9 digits now invalid, shortest national number length now 10, so parsing affected for numbers shorter than this)
IS (8 digits now invalid)
KG (7,8 digits now invalid)
KR (11 digits now invalid)
LA (7 digits now invalid)
LI (8 digits now invalid)
LY (8 digits now invalid)
MV (8 and 9 digits now invalid)
MW (8 digits now invalid)
MX (9 digits now invalid)
NP (9 digits now invalid)
SE (11 digits now invalid)
SG (9 digits now invalid)
SL (7 digits now invalid)
SM (7-9 digits now invalid)
UA (8 digits now invalid)
UG (8 digits now invalid)
UZ (8 digits now invalid)

- LOGGER -> logger
- Small comment fixes
- Style-guide-recommended formatting changes
- Removal of duplicate, unneeded method in PhoneNumberToTimeZonesMapper.java for canBeGeocoded, using phoneNumberUtil instead.

We remove the dependency on protobuf (protobuf-javanano) from the Java library, resurrecting java/libphonenumber/src/com/google/i18n/phonenumbers/Phonemetadata.java which was present until libphonenumber-7.1.1.

Does not affect the C++ or JavaScript libraries.

https://groups.google.com/forum/#!topic/libphonenumber-discuss/75TOpTFVi08

* Initial code changes to support the new possibleLengths infrastructure
when building metadata. Includes setting these in the main metadata
file and changing the code to populate them in the PhoneMetadata proto
at build time. Updating tests for Java.

* Updating comment about the possible_lengths_local_only field.

* Adding the generated jars that build files containing the new possible
lengths fields.

* Rebuilding short-number metadata with the new proto field. Not doing
this at release time so we can see what changes in the metadata are due
to metadata fixes, rather than this possibleLengths field.

* Regeneration of phone number metadata with the new possible length
information filled in.

Includes some possible length changes for KR and BY, and a validation
fix for NA where a digit was missing.

* Updating the test proto metadata files with the new possible length
information.

* Added notes about the code changes.

* Regenerating C++ metadata with new possibleLength info and updating unit
test to check this works.

Making function IsNumberGeographical public. This function operates on
the type of number and the country it belongs to only, so may have some
false positives.

Using this in the geocoder so that geocoding now limited to numbers that
we consider geographical, based on their type and country, rather than
just based on their type. The C++ geocoder did not previously check the
number type/country at all.

Indonesian and Chinese mobile numbers have now been added to the list of
possibly-geographical numbers.

* Metadata updates for release 7.3.2

* Unit tests for Canada Short number updates

example number for every region, and that we have an example number for
every type. Also adding better logging for example short number cost
mismatches.

* New useful getExampleNumber methods (for testing) - one for invalid
  numbers and one for getting a number of a particular type without
  specifying a country.

MultiFileMetadataSourceImpl.

... with per-region files' prefix; see:
- https://github.com/googlei18n/libphonenumber/blob/master/java/libphonenumber/src/com/google/i18n/phonenumbers/MultiFileMetadataSourceImpl.java
- https://github.com/googlei18n/libphonenumber/blob/master/java/libphonenumber/test/com/google/i18n/phonenumbers/TestMetadataTestCase.java