Add a first file for a driver for the Keysight 34465A DMM and a
benchmarking notebook

* Fix: Partial fix for microsoft#499 propagate units to subarray

* Fix: This is a setpoint array by definition

* fix: test_loop rename setpoint arrays

* fix: test_measure rename setpoint arrays

* Fix: Add setpoint_units to Array and Multiparameter

* Fix: add handling of setpoint units to loops

* Add test for multiparameter units setpoint_units

* Remove assert

* Add setpoint_units to meta_arrr so its correctly shapshoted

* Fix: remove unused variable

* Initial draft of the DataSet spec

* Updated based on feedback

* More feedback

Removed API reference to QCoDeS Parameters, allowed addition of
parameters with value arrays, dropped "optional", added min_count and
min_wait to subscriptions, and some RST cleanup.

* Added unique identifier

Added a name parameter to the constructor and an id attribute that
returns a unique identifier for the DAtaSet, suitable for use as a
reference.

* Id field should be part of the metadata

Specified that the identifier should be automatically stored in the
DataSet's metadata.

* Nearly-final draft

Removed persistence, added more metadata details, added utilitiy
function section

clean-up of PR microsoft#235. I suspect buggy behaviour but have not tested the driver.

…soft#521)

Closes microsoft#519

Make it possible to not query the currents on startup. Also, do NOT set the voltage range during init, since this could cause voltages to jump.

Make channels be always bound to a unique function generator.

Add a unit to voltage ramp slope.

* add driver for HP-83650A

* autopep

* remove doc

* some fixes from PR comments

* rename units to unit

* fix all pull request review comments

* autopep8

* fix changes made in (e1d9ece)

* change units to unit in print methods

* remove __name__ from debug

* feat: Add a driver for ZI UHF-LI

First commit providing a driver with a working sweeper. Needs
refactorising the sweeper into a MultiParameter

* fix: Change the sweep into a MultiParameter

Change the sweep from a normal instrument method into a MultiParameter

Breaking Changes: execute_sweep no longer exists. All sweep data goes
through the Sweep parameter.

* fix: Tidy code

Tidy code and remove whitespace

* feat: Add example notebook

Add an example notebook explaining how to use the sweeper.

* fix: Correct multiple dict keys

Correct the multiply-defined dict keys of the demodulator trigger.

* updating ZIUHFLI.py on dummy setup

* intermediate commit

* testing"
"

* feat: Add parameters

Finish Sigout parameters and start adding scope parameters.

* Intermediate commit

To be cleaned later

* feat: Add scope get function

Add a get function to the scope multiparameter and document this in the
notebook.

* fix: Update setpoints with delay

Update the scope setpoints with reference and delay values.

* fix: Adjust setpoints

Adjust the setpoints of the scope to match the expectations of the loop.

* fix: pep8

Break lines and trim whitespace.

* fix: Add scope instance to instrument

Add a scope instance to the __init__ of the instrument.

* docs: Fix requested changes

Add requested docs

Driver/QDac

* feat: Add a driver for Keysight 33500B

Add a driver and a notebook.

* feat: Add phase parameter

Add a parameter controlling the output phase

Adds clock_source and ref_source parameters
Also added a change to the get parser of the lowpass filter setting. The
value 9.9e37 corresponds to infinity but can give problems when sending
it over a port (such as to pyqtgraph).

Awg enhancements

* remove metaclass

* first step in removing the metaclass

* fix: Remove background

* fix: remove alt-docs about monitor

* remove signal queue

* fix: More backgroudn removal

* remove data_manager stuff

* feature: Remove tests, yolo

* feature: Rremove more mocks

* finally eradicate mp

* remove Widgets

* chore: Remove benchmarking and reloading

Benchmarking should go if anything to its own repo
reloading is dangerous, and it can trick many new users.

* chore: Add todo

* chore: Use py.test

move faster without reinventing the wheel at all times

* fix: Remove nested attrs as unused

* Fix remove try/Interrupt, missing variable imax

* fix: Remove trailing number from test name

* fix: Use unit

* chore: Update documentation

* fix: Remove toymodel, and simlify examples

* docs: Remove more MP from docs

* chore: Remove untested example adaptive sweep

* docs: Update docs

* fix: remove mp from config

* fix: Remove server references

* chore: Add todo

* fix: Remove server and metaclass

* add stuff

* fix: use right default for _instances

* bugs

* feature: add more tests

* Remove find_component

* fix vim typo

* fix: Remove array getter

* fix: remove timing

* feature: Allow for legacy code to run

This allows for no exception but just a warning.
The deprecated argument is ignored.

* Fix: zi ensure single trigger mode is always set

This get reset when enabling the read

* fix move trigger mode setting to where it belongs

* Fix: zi reuse scope and handle errors by retrying

* Fix: typo

Keysight add missing init file

ZI: clear history when reusing scope

[WIP] Zi fixes

We need the phase parameter for fast charge diagrams

Add a first file for a driver for the Keysight 34465A DMM and a
benchmarking notebook

Add four trigger parameters to the driver.

Make a more general get_cmd for the volt parameter. Using simply 'READ?' is wrong in many cases.

Add four parameters for sampling.

…codes into driver/keidmm