[telemetry] Add support for BenchmarkResults that vary from page to page

[telemetry] Add support for BenchmarkResults that vary from page to page
This patch started out trying to add support for benchmarks that had different
outputs for each page. This is required for benchmarks that report [filtered]
profiling data, for instance something that simply groups timeline data and
reports it. To do csv output in taht situation, we have to hold output until
the end of the benchmark.

Making this work cleanly in the multi page benchmark code turned out to be
easily hacked, but unearthed a pretty hefty amount of deep, branchy code. So,
to make this cleaner, I cleaned up the results object.

We now have:
 - PageBenchmarkResults: all results for the benchmark
 - ValuesForSinglePage: all the results for a page
 - PageBenchmarkValues: a single value from a benchmark

PerfBot benchmark output is managed by PageBenchmarkResults. Csv and Block

terminal-block was renamed to block, but with backcompat to terminal-block.
Nobody could remember terminal-block.

R=tonyg,hartmanng
NOTRY=True

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=175602

[hartmanng, 2013-01-08 16:04:10]

LGTM with nit

https://codereview.chromium.org/11779041/diff/3001/tools/telemetry/telemetry/...
File tools/telemetry/telemetry/multi_page_benchmark_runner.py (right):

https://codereview.chromium.org/11779041/diff/3001/tools/telemetry/telemetry/...
tools/telemetry/telemetry/multi_page_benchmark_runner.py:18: from telemetry
import block_page_benchmark_results
The rest of these imports look to be alphabetical, might as well keep it that
way.

[tonyg, 2013-01-08 17:07:08]

LGTM too. Beautiful cleanup!

[commit-bot: I haz the power, 2013-01-08 21:14:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nduca@chromium.org/11779041/5001

[commit-bot: I haz the power, 2013-01-08 22:18:48]

Change committed as 175602