Add an --in switch to ipc_message_util to support minimization in CF.

Add an --in switch to ipc_message_util to support minimization in CF.

BUG=450268
R=inferno@chromium.org,tsepez@chromium.org

Committed: https://crrev.com/66c7eff8184adfdcd58e2d381956382635b17ecf
Cr-Commit-Position: refs/heads/master@{#316593}

[inferno, 2015-02-17 16:53:01]

The CQ bit was checked by inferno@chromium.org

[inferno, 2015-02-17 16:53:02]

lgtm

[commit-bot: I haz the power, 2015-02-17 16:53:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/931153003/1

[commit-bot: I haz the power, 2015-02-17 17:13:05]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-02-17 17:16:10]

Patchset 1 (id:??) landed as
https://crrev.com/66c7eff8184adfdcd58e2d381956382635b17ecf
Cr-Commit-Position: refs/heads/master@{#316593}

[Tom Sepez, 2015-02-17 18:50:47]

https://codereview.chromium.org/931153003/diff/1/tools/ipc_fuzzer/mutate/mess...
File tools/ipc_fuzzer/mutate/message_util.cc (right):

https://codereview.chromium.org/931153003/diff/1/tools/ipc_fuzzer/mutate/mess...
tools/ipc_fuzzer/mutate/message_util.cc:33: "output only the messages in the
specified list.";
nit: is this a list of message IDs or a list of ordinal positions?

https://codereview.chromium.org/931153003/diff/1/tools/ipc_fuzzer/mutate/mess...
tools/ipc_fuzzer/mutate/message_util.cc:105: bool has_indices =
cmd->HasSwitch(kInSwitch);
nit: You can move these down to line 137 or so since it doesn't look like they
are used earlier.

https://codereview.chromium.org/931153003/diff/1/tools/ipc_fuzzer/mutate/mess...
tools/ipc_fuzzer/mutate/message_util.cc:106: std::vector<bool> indices;
nit: you probably want std::set<int> here.  If I'm reading this right, and the
vector is indexed by the message id, the storage could be large due to the
sparse nature of the way we generate message IDs.  On the other hand, if the
vector is indexed by the ordinal position in the message file, this is  OK.