Issue futurewarning

[jaclark5]

Fixes #408, by adapting code to resolve pandas FutureWarnings:

 [alchemlyb/src/alchemlyb/estimators/bar_.py:100](alchemlyb/src/alchemlyb/estimators/bar_.py:100): FutureWarning: When grouping with a length-1 list-like, you will need to pass a length-1 tuple to get_group in a future version of pandas. Pass `(name,)` instead of `name` to silence this warning.
  (len(groups.get_group(i)) if i in groups.groups else 0)
[alchemlyb/src/alchemlyb/estimators/bar_.py:122](alchemlyb/src/alchemlyb/estimators/bar_.py:122): FutureWarning: When grouping with a length-1 list-like, you will need to pass a length-1 tuple to get_group in a future version of pandas. Pass `(name,)` instead of `name` to silence this warning.
  uk = groups.get_group(self._states_[k])
[alchemlyb/src/alchemlyb/estimators/bar_.py:127](alchemlyb/src/alchemlyb/estimators/bar_.py:127): FutureWarning: When grouping with a length-1 list-like, you will need to pass a length-1 tuple to get_group in a future version of pandas. Pass `(name,)` instead of `name` to silence this warning.
  uk1 = groups.get_group(self._states_[k + 1])

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.78%. Comparing base (aba3671) to head (0b718ea).
Report is 8 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #407   +/-   ##
=======================================
  Coverage   98.78%   98.78%           
=======================================
  Files          28       28           
  Lines        1978     1978           
  Branches      435      435           
=======================================
  Hits         1954     1954           
  Misses          2        2           
  Partials       22       22           

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

[orbeckst]

thanks for the fix.

(looks a bit uglier than it used to but if this what pandas wants...)

[orbeckst]

Let's include this fix in the 2.4.1 release #403

[jaclark5]

thanks for the fix.

(looks a bit uglier than it used to but if this what pandas wants...)

@orbeckst The alternative is to have all lambda states be represented as tuples. so instead of:
u_nk.columns.tolist()
> [0, 0.25, 0.5, 0.75, 1.0]
it will have to be:
u_nk.columns.tolist()
> [(0,), (0.25,), (0.5,), (0.75,), (1.0,)]

[orbeckst]

Thanks @jaclark5 !