Restructure for Micromanager 2.0 Gamma

[camFoltz]

I have started a new branch that will begin a restructure of the current reconstruct-order metadata/workflow. Instead of relying on reading metadata files, it will use a reader that @ieivanov wrote found here: https://github.com/czbiohub/aicsimageio/tree/feature/micromanager_reader/aicsimageio

The new branch can be found at: mm2_restructure

it requires the installation of Ivan's reader, which can be done by:
pip install git+https://github.com/czbiohub/aicsimageio.git@feature/micromanager_reader

What this will do is allow us to scrap the existing metadata structure, which dominates how the entire workflow is ran. Specifically, it uses parts of the metadata file that may get miswritten/not correspond to the actual names of the images, and it becomes a nuissance to try and hack it to work. In this restructure, we aim to use this simple reader, in which we can call any image arrays from any specific dimension of a large tiff_stack. This will allow us to collect data from MM2.0 as tiff stacks and not worry about all of the tiny naming conventions.

I have begun to implement this reader, starting with reconstructBatch.py . I suggest following my format of copying the original python file and creating a new one to involve the new reader, titled reconstructBatch_MM2.py for example. This will allow for easy comparison and for the ability to keep backwards compatibility.

The goal would be to rework everything that is reliant on old metadata structure, and replace with calling a metadata object created from AICSImage. This will simplify the current structure dramatically, as well as allow us to start using MM2.0 data reliably.

I would love some help, my goal for the next bit will be modifying multiDimProcess.py and moving on to the dependencies from there.