Add a C sample defining a large initialized array

[ib]

This fixes github issue #5012.

• I am fixing a misclassified language
  • I have included a new sample for the misclassified language:
    • Sample source(s): C/foo.h
      • [URL to each sample source, if applicable]
    • Sample license(s): file created by me, any license you want
  • I have included a change to the heuristics to distinguish my language from others using the same extension.

[lildude]

🤔 I'm not sure this is really solving the real issue here. The classifier uses real-world samples to try and guess the language based on how it is really used. This contrived sample is being added to deliberately sway the classifier and in turn may start causing legitimate Objective-C or C++ header files (I have no idea if there is such a thing) to be classified as C and thus only add to what is already a tricky problem as detailed in the ongoing discussion in #1626

[ib]

The sample is based on real-world headers: https://github.com/ib/gucharmap/tree/master/gucharmap: unicode-*.h

[lildude]

The sample is based on real-world headers: https://github.com/ib/gucharmap/tree/master/gucharmap: unicode-*.h

Why not use one of the real-world files?

[smola]

The main problem here is that token frequency within documents leads to weird results in classification. For example, if a few more braces appear in C++ files, the classifier will be biased towards C++ when classifying pure C files that have many braces.

I think the right solution is going for the root problem: absolute frequency of tokens is suboptimal, and we should probably move to document frequency. I've been testing this approach and I think it should work decently (and faster). But it still requires some more verification.

[stale]

This pull request has been automatically marked as stale because it has not had recent activity, and will be closed if no further activity occurs. If this pull request was overlooked, forgotten, or should remain open for any other reason, please reply here to call attention to it and remove the stale status. Thank you for your contributions.

[lildude]

Closing as this isn't a real world sample and is only being added to improperly influence the classifier. Recent and pending improvements to the classifier will also help towards addressing the original issue.