Add generic interactive shell

[shihab-dls]

fixes #140

This PR adds an interactive shell, based on this implementation, and adds transport contexts. The wip commits are attempts to add dbl dbpf and dbgf functionality to PVA as an example, but maybe only the first commit is required for this PR, and further PRs can add transport specific context.

[shihab-dls]

I will request an initial review to decide on steps forward (i.e., reverting wip commits and keeping generic, or continuing on transport context).

The failed test is due to asserting a specific message is printing out, where the terminal is now injecting it's messages (e.g., 'In [1]:). Should we disable the terminal with some test ENV VAR instead of working around it in these sorts of tests?

[shihab-dls]

I've removed implementations of transport context for pva; instead I made this #190 for CA. This PR is now only the generic implementation.

[GDYendell]

Some small comments

[codecov]

Codecov Report

❌ Patch coverage is 89.28571% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 92.20%. Comparing base (64a7828) to head (e72f1f5).

Files with missing lines	Patch %	Lines
src/fastcs/launch.py	88.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #189      +/-   ##
==========================================
- Coverage   92.25%   92.20%   -0.06%     
==========================================
  Files          40       40              
  Lines        2040     2065      +25     
==========================================
+ Hits         1882     1904      +22     
- Misses        158      161       +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[shihab-dls]

I've commented out fastcs.run() lines that were in some doc snippets for now, and removed the env var. Re-requesting a review with the generic terminal and the amended test. I've also added a test for the duplicate context error for coverage; let me know if you would've tested it another way.

[GDYendell]

I think I am OK with leaving that run method with missing lines unless you can come up with a neat way to test that

[shihab-dls]

I think I am OK with leaving that run method with missing lines unless you can come up with a neat way to test that

Hit my head against it for a little while, and can't figure out a "neat" way to do it; I can open up a small issue for me to look at after, but for now we should probably merge this in (if you are happy with my other test)?