Fix 1679 and 1681

[jajhall]

Renamed

• addVar to addVariable to prevent shadowing of highspy binding to Highs::addVar
• Change the name of removeVar to deleteVariable for consistency with Highs::deleteCol
• Change the name of getVars to getVariables

Added a call of self.update() to addVariable, rather than deferring it to addConstr, to allow column data to be changed, and rows to be added, using highspy bindings to methods in the Highs class. These will assume that all variables that have been defined are in the incumbent model.

Unfortunately - see experiments with examples/distillation.py - this gets the column names wrong. It would appear as if the names stack isn't flushed when self.update() is called. Any thoughts @galabovaa ?

[jajhall]

This will close #1679 and #1681

[galabovaa]

OK it does seem that names are not cleared, this seems to be by design because the idea I think was to call update() once all the modelling is done, just before run.

I tried to clear them naively at the end of update() in branch fixx, but that breaks one test:

TestHighsPy::test_deleteVariable FAILED
https://github.com/ERGO-Code/HiGHS/actions/runs/9051630828/job/24868361542

Here is the failing test code:
`
def test_deleteVariable(self):
h = highspy.Highs()
x = [h.addVariable(), h.addVariable()]

    h.update()
    self.assertEqual(h.numVars, 2)

    h.deleteVariable(x[0])
    self.assertEqual(h.numVars, 1)

`

Natually, once I have cleared the list, the modelling language can not find the variable to delete.

I'll investigate further

[jajhall]

'update()' is called when each constraint is added - otherwise addRow() returns kHighsError.

It seems as if just the names fail - possibly something Luke didn't test.

[galabovaa]

I added a list of names to the batch and now call that with names: see the latest commit.

I thought it would be easier than trying to keep track of batch indices in vars_. Now the batch names are cleared at the end of update() and the column names should look allright

** edit something is not right again because there is a test_var_name failing, will look further

[jajhall]

Yes, this fixes it. It's actually nothing to do with adding update() to addVariables, since this would be wrong once update() has been called for the first non-empty batch.

[jajhall]

something is not right again because there is a test_var_name failing, will look further

We now can't change name thus

# change name before adding to the model
    x.name = 'y'
    self.assertEqual(x.name, 'y')

    # add to the model
    h.update()

TBH, I can't see a value in allowing this to be possible, so we can delete the test