Adds feature to export positioner pose from a simulation

[StephenNneji]

• Please check if the PR fulfills these requirements

• The commit message follows our guidelines
• Tests for the changes have been added (for bug fixes / features)
• Docs have been added / updated (for bug fixes / features)

• What is the new behaviour (if this is a feature change)?
  This adds the option to download positioner pose (position and orientation) matrices after a simulation. To use the feature:

1. Run a simulation as normal.
2. When the simulation is done, click the download button shown in the image below.
3. Select save location
4. Each row in the file is a flatten matrix. The matrices are stored column first i.e. [m00, m10, m20, m01, m11, m21, m02, m12, m22, m03, m13, m23] so the first 9 values are for the rotation matrix and the last 3 the translation vector.

• Does this PR introduce a breaking change (What changes might users need to make in their application due to this PR)?
  No

• Other information:
  

[andy-bridger]

Seems to work, if I could be cheeky and request comma separated values in the output file, that would be lovely, but can obviously process the string on our end if needed.

I haven't done I functional test of the matrix being output (although I trust it is correct, I just mean testing it does the same thing in mantid) but that might have to wait until we collect the new experimental data at the end of May (when we have the actual physical reference)

[StephenNneji]

Seems to work, if I could be cheeky and request comma separated values in the output file, that would be lovely, but can obviously process the string on our end if needed.

No problem, I will make the output comma separated

[RabiyaF]

I tested it and it works as expected. I just wanted to mention one issue. I can't create new environments for sscanss. I get the following error on M1 and M3 macs.

I have some old environments for fitbenchmarking that have Nlopt==2.7.1 installed. I used one of these old conda environments to test the new features of this PR and they work as expected.

Maybe it might be worthwhile to consider updating the numpy dependency to > 2.x.x which will allow installing Nlopt > 2.7.1.

I briefly looked into this. It can be done very easily using ruff. There might be ~13 lines that need to be updated in the codebase. They are the lines that use row_stack.

[StephenNneji]

Maybe it might be worthwhile to consider updating the numpy dependency to > 2.x.x which will allow installing Nlopt > 2.7.1

I will do another PR for upgrading package versions

[StephenNneji]

Seems to work, if I could be cheeky and request comma separated values in the output file, that would be lovely, but can obviously process the string on our end if needed.

@andy-bridger, I had a look at the other exports and tabs are used everywhere, so to keep consistency I will leave the tab separated values as is, but I could revisit in the future maybe by adding a setting to allow the delimiter changes globally.