Netpyne

- Hotfix to support new GUI functionality (fixes in grid parameter optimization)

- Removed redundant warnings

**New features**

- Added ability to load PointCell from saved network

**Bug fixes**

- Fixed bug with loading CompartCell with custom mechanisms from saved network

- Fixed bug with accessing proper sim data file during batch parameters optimization

- Fixed bug where most batch optimization methods used to misname simulation output file

- Fixed bug in pop.py: "gridSpacing" is in the list of excludeTags

- Fixed bug in network.py: in the list "stimStringFuncParams" --> 'del' instead of 'delay', and removed 'i'

- Fixed bug in conn.py: included "params['preSec'] = connParam.get('preSec')" in gapJunctions

- Fixed bug in stim.py: 'Loc' with capital letter, for making 'synMechLocFactor'

- Fixed some misinformation in reference.rst about the subconn

- Fixed bug in dipole calculation units - changed from mA to uA

- Fixed bug in conditional logic when gathering LFP / dipoles

- Allow tuples to specify population's cells in 'include' for plotSpike

- Cleaned up optuna optim tutorial

**New features**

- Added axis sharing and twinning capability

- Updated metadata with dipole specs

- Updated some LFP examples

- Added clearAll argument to sim.create()

- Changed ordering of documentation to “by source”

**Bug fixes**

- Fixed bug in plotting when overlaying axis

- Fixed bug with syncLines option in plotter

- Fixed bugs in MultiFigs (2D subplots, bug in LFP with multi axis)

- Fixed bug in checkAvailablePlots()

- Fixed gathering dipoleSum and LFPPops data from nodes

**New features**

- Added back the syncLines argument to plotRaster()

- Updated conversion to/from NeuroML 2

- Added RxD-based spreading depression example model

**Bug fixes**

- Avoid exception in plotSpikeHist() and plotSpikeFreq() when pop has no spikes

**New features**

- Added new plotting module with custom plotters

- Separated analysis.plotSpikeRaster into analysis.prepareRaster and plotting.plotRaster

- Separated analysis.plotSpikeHist into analysis.prepareSpikeData, analysis.prepareSpikeHist, plotting.plotSpikeHist and plotting.plotSpikeFreq

- Separated analysis.plotLFP into analysis.prepareLFP, analysis.preparePSD, analysis.prepareSpectrogram, plotting.plotLFPTimeSeries, plotting.plotLFPPSD, plotting.plotLFPSpectrogram, and plotting.plotLFPLocations

- Mapped old analysis.plots onto new ones

- Interfaced with LFPkit to enable recording/plotting of dipole current moments and EEG signals

- Added Current Source Density (CSD) analysis and plots

- Enabled selecting a subset of cells to save individual LFP signal from via cfg.saveLFPCells = [...]

- Added cfg.saveLFPPops to store LFP generated individually by each population

- Function to return batch parameter combinations (used for GUI)

- Added jupyter nb tutorials

- Added hippocampus CA3 model example

- Updated SONATA importing

- Improved distributed saving and interval saving (more robust)

- Added mapping of netParams values within cfg (used for batch sims in GUI)

- Extended Action tests to pull requests (not just push)


**Bug fixes**

- Fixed setting cfgMapping nested params when not string

- Fixed defineCellShapes when sec['hObj'] is not a h.Section()

- Fixed bug when using lognormal in string functions

- Added rxdmath to avoid RxD error

- Fixed bug in Tutorial 8 (batch)

- When saving to JSON wait until file exists before returning (to avoid exiting before saving finished)

- Release for use with GUI