Pyerrors

Added
- Alias `gm` for `Obs.gamma_method` added.
- Hotelling t-squared p-value added for correlated fits.
- String conversion of numpy arrays containing `Obs` improved.
- Input routine for xSF measurement program added.

Fixed
- Complex valued `Corr` objects fixed.
- Small bug in `qtop_projection` fixed.
- Bug in `Corr.spaghetti_plot` fixed which appeared in connection with replica separators.

Changed
- Merged `Obs` are no longer filtered as this lead to inconsistent `idl`s in some edge cases. Error estimates are unaffected up to filter precision.

Removed
- Removed the `Obs` attribute `is_merged` as this information was only needed for the filtering. The change results in a ~1.15x speed up in the multiplication of two `Obs`.

Added
- Log-derivatives and symmetric log-effective mass added.
- Covariance for irregular Monte Carlo chains sped up.
- Additional checks in `Corr.GEVP` added.

Fixed
- Bug in `Obs.details` fixed which appeared when tau had zero error.
- Bug in `input.json` export in connection with `numpy.int64` fixed.
- Small bug fixes in `input.openQCD`.

Fixed
- Integrated autocorrelation times are now correctly estimated for gapped irregular Monte Carlo chains.
- The output of `Obs.details` was improved and now contains information about the stepsize in configurations for which the integrated autocorrelation time was estimated.

Added
- `least_squares` and `total_least_squares` fits now have an optional keyword argument `num_grad`. If this argument is set to `True` the error propagation of the fit is performed via numerical instead of automatic differentiation. This options allows for fits functions which contain special functions or which are not analytically known.

Fixed
- Bug in `Corr.show` `comp` option fixed.

Added
- New submodule `input.pandas` added which adds the possibility to read and write pandas DataFrames containing `Obs` or `Corr` objects to csv files or SQLite databases.
- `hash` method for `Obs` objects added.
- `Obs.reweight` method added in analogy to `Corr.reweight` which allows for a more convenient reweighting of individual observables.
- `Corr.show` now has the additional argument `title` which allows to add a title to the figure. Figures are now saved with `bbox_inches='tight'`.
- Function for the extraction of the gradient flow coupling added (see 1607.06423 for details).
- `Corr.is_matrix_symmetric` added which efficiently checks whether a correlator matrix is symmetric. This is used to speed up the GEVP method.

Fixed
- `Corr.m_eff` can now deal with correlator entries which are exactly zero.
- Minor bugs in `input.dobs` fixed.

Fixed
- Further bugs in connection with correlator objects which have arrays with None entries as content fixed.