an inappropriate @strict_mode_negative to @noStrict. There are more
(many more?) of these to fix

these tests were valid tests of best practices, but not of the
normative spec. We moved these to a new bestPractices directory and
added a new not-yet-operational @bestPractices property.

Added new 'packaging' directory to tooling.

Generate the 'id' property from the 'path' property.

Added all tests added to repairES5.js since the last time I contributed SES-relevant tests to test262.

(between David Fugate and I) canonical test262 test format

- juneDate and decemberDate were both one month off.  Fixed
- LocalTZA wasn't actually LocalTZA as defined in ES5.  If we were in DST when
  this was created, LocalTZA was off by one hour.  Fixed
- GetSundayInMonth was completely busted (i.e., arithmetic operations on the 'count'
  param which happens to be a string).  Fixed(?)
- DaylightSzavingTA was defined incorrectly.  It assumed the local time zone adjustment
  hadn't been accounted for.  This wasn't how UTC was calling it.  Fixed(?)

failed.

applied to many fixed sources, corresponding to previous two commits

gratuitous strict mode incompatibility. Too many to continue fixing
manually this way. Don't know what to do about that.