Support for URLs analysis in cuckoo distributed (no db migration needed)

[fyhertz]

Usage example:
curl http://localhost:9003/api/task -F url="http://www.perdu.com"

I think this is a better way to implement this feature than my previous pull request as no database migration is needed. The path field is used to store URLs in the task table.

[coveralls]

Coverage decreased (-0.03%) to 58.544% when pulling 8e96c3b on fyhertz:urldistributed2 into 2611ef6 on cuckoosandbox:master.

[codecov-io]

Codecov Report

Merging #1700 into master will decrease coverage by 0.03%.
The diff coverage is 32%.

@@            Coverage Diff             @@
##           master    #1700      +/-   ##
==========================================
- Coverage   58.57%   58.54%   -0.04%     
==========================================
  Files         147      147              
  Lines       14094    14104      +10     
==========================================
+ Hits         8256     8257       +1     
- Misses       5838     5847       +9

Impacted Files	Coverage Δ
cuckoo/distributed/views/api.py	14.13% <0%> (-0.24%)	⬇️
cuckoo/distributed/api.py	80.35% <57.14%> (-9.44%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2611ef6...8e96c3b. Read the comment docs.

[jbremer]

Although I mostly agree that this PR supersedes #1694, I somewhat also feel like adding a category field - much like we have in the Cuckoo Core - would make the changes in this PR slightly less hacky.
What are your thoughts?

[fyhertz]

It would be less hacky indeed! In the meantime, how about merging this cheap PR to enable distributed URLs analysis? Another later commit could add the category field and start using it to differentiate samples from URLs (instead of checking if filename is empty) while the path field would still be used for storing the URLs.

In other words, I think this PR brings distributed closer to a clean solution anyway. What do you think?

[jbremer]

If I've learnt one thing then it's the fact that such 'temporary' fixes remain permanent ;-) As we have to maintain it it's much preferable to get it right the first time.
That said, and as you agreed the other option would be slightly better, would you be up for implementing it? Thanks!

[jbremer]

I guess the answer is 'no' :-) We'll probably be implementing this ourself eventually (pinging @RicoVZ), closing PR. Feel free to reopen if you'd like to implement the proper approach, as discussed above.
Ah, just saw the other PR, we can do the category column addition on top of that PR.