Add Mojo IPC for seeding new Renderer with Browser's cached blob state.

blimp/common/blob_cache/blob_cache.h

Index: blimp/common/blob_cache/blob_cache.h
diff --git a/blimp/common/blob_cache/blob_cache.h b/blimp/common/blob_cache/blob_cache.h
index 1f8a2932723640f2346321797ca98e29cbcb9f5d..b1c7118f102510495fc5a25257c989de31bb07a1 100644
--- a/blimp/common/blob_cache/blob_cache.h
+++ b/blimp/common/blob_cache/blob_cache.h
@@ -25,6 +25,9 @@ class BLIMP_COMMON_EXPORT BlobCache {
  public:
   virtual ~BlobCache() {}
 
+  // Gets the keys of items stored by the cache.
+  virtual std::vector<BlobId> GetCacheState() const = 0;

[Wez, 2016/06/21 00:33:33]

nit: GetCachedBlobIds()?

[Wez, 2016/06/21 00:33:33]

In general this seems like a potentially large structure to be generating and
returning?

[Kevin M, 2016/06/21 21:23:51]

Done.

[Kevin M, 2016/06/21 21:23:51]

Agreed, but in general I think if this saves us even a few image encodes, the
benefits might outweigh the cost?

I think we can do away with this approach once we get asynchronous image loading
in Skia, since renderer will be able to directly query the browser's cache state
using Contains() and postpone the encode/IPC until it receives a negative
result.

[Wez, 2016/06/22 19:54:23]

Acknowledged.

+
   // Returns true if there is a cache item |id|.
   virtual bool Contains(const BlobId& id) const = 0;