factory_test_tools: use flimflam to get connection info directly.

factory_test_tools: use flimflam to get connection info directly.

To  preven using "modem status" command, this CL tries to load flimflam
tests directly.

BUG=chromium-os:13967, chrome-os-partner:2952
TEST=./gft_hwcomp # seeing:
  'part_id_3g': ['0XXX:XXXX Some Manufaturuer Some Product'],
  'version_3g_firmware': ['Some Vendor 09010012'],

Change-Id: I648a1afcfe8a319f4bc3f5c981c12048d0648235

R=yongjaek@chromium.org,jglasgow@chromium.org

Committed: http://chrome-svn/viewvc/chromeos?view=rev&revision=1b50ff1

[Hung-Te, 2011-04-13 18:07:23]

NOTE: the format and content for "3g firmware" may still be changed before
commiting.

I think current R11 implementation is weird because it uses the "Version" from
GetInfo, which is in fact "HardareRevision" in Gobi2k.

Maybe the real "firmware version" and "carrier information" is helpful?

[yongjaek, 2011-04-13 18:45:01]

R11 is based Y3300, and we got a confirmation from Samsung that Version is the
firmware info.

[Hung-Te, 2011-04-14 04:27:57]

I've updated the generation rule so it will try 
 Carrier + Firmware Revision
 Firmware Revision
 Hardware Revision
in order.

Please take a look again, thanks.

[Hung-Te, 2011-04-15 11:11:54]

Any feedbacks? thanks

[Jason Glasgow, 2011-04-15 22:39:44]

http://codereview.chromium.org/6835028/diff/5001/gft_hwcomp.py
File gft_hwcomp.py (right):

http://codereview.chromium.org/6835028/diff/5001/gft_hwcomp.py#newcode601
gft_hwcomp.py:601: (_, attributes) = self.load_flimflam()
Looks okay to me.  I would have used mm.py and gone directly to the modem
instead of going through flimflam, but I guess this works.