Move cmor_check_data to early in preprocessing chain

[bouweandela]

Moving the CMOR check of the data to earlier in the default preprocessor order, will make it possible to implement preprocessor functions that change the units of the data without using a custom_order preprocessor. People seem to run into this again and again.

Related to #708, #442, #60

Closes #741, closes #212

Tasks

• Create an issue to discuss what you are going to do, if you haven't done so already (and add the link at the bottom)
• This pull request has a descriptive title that can be used in a changelog
• Add unit tests
• Circle/CI tests pass. Status can be seen below your pull request. If the tests are failing, click the link to find out why.
• Codacy code quality checks pass. Status can be seen below your pull request. If there is an error, click the link to find out why. If you suspect Codacy may be wrong, please ask by commenting.

---

[hb326]

I ran my recipe again that gave me problems in the earlier test (#727) with the new order of preprocessing. It ran through, but the units were still not correct. For the anomaly plot I would expect something with unit=1, but I get the unit of the variable that I use in the recipe.

[bascrezee]

@hb326 Since the units are set to 1 in another branch ( std-anoms-are-unitless) you have to use that one to see effect. I now merged preproc-check-order such that you can check the combined behaviour. Can you try it again? With the branch std-anoms-are-unitless? @bouweandela Thanks!

[bascrezee]

Succesfully tested in combination with #727.

[bascrezee]

Looks good to me!

[jvegreg]

Already tested in the anomalies branch