Pyfixest

- Fixes a bug in `etable()` and a warning in `fixest_model_matrix` that arose with higher `pandas` versions. Thanks to aeturrell for reporting!

New Features

- Introduces a new [pyfixest.did](https://github.com/s3alfisc/pyfixest/tree/master/pyfixest/did) module which contains routines for Difference-in-Differences estimation.
- Introduces support for basic versions of the local projections DiD estimator following [Dube et al (2023)](https://www.nber.org/papers/w31184).
- Adds a new [vignette ](https://s3alfisc.github.io/pyfixest/difference-in-differences-estimation/)for Difference-in-Differences estimation.
- Reports R2 values in `etable()`.

Enhancements:

- Good performance improvements for singleton fixed effects detection. Thanks to [styfenschaer](https://github.com/styfenschaer) for the PR! See [#229](https://github.com/s3alfisc/pyfixest/issues/229).
- Uses the [r2u project](https://github.com/eddelbuettel/r2u) for installing R and R packages on github actions, with great performance improvements.
- Allows to pass `polars` data frames to `feols()`, `fepois()` and `predict()`. [232](https://github.com/s3alfisc/pyfixest/issues/232). Thanks to [vincentarelbundock](https://github.com/s3alfisc/pyfixest/issues/232) for the suggestion!

Bug Fixes:

- Missing variables in features were not always handled correctly in `predict()` with `newdata` not `None` in the presence of missing data, which would lead to an error. See [246](https://github.com/s3alfisc/pyfixest/issues/246) for details.
- Categorical variables were not always handled correctly in `predict()` with `newdata` not `None`, because the number of fixed effects levels in `newdata` might be smaller than in `data`. In consequence, some levels were not found, which lead to an error. See [245](https://github.com/s3alfisc/pyfixest/issues/245) for details. Thanks to [jiafengkevinchen](https://github.com/jiafengkevinchen) for the pointer!
- Multicollinearity checks for over-identified IV was not implemented correctly, which lead to a dimension error. See [236](https://github.com/s3alfisc/pyfixest/issues/236) for details. Thanks to [jiafengkevinchen](https://github.com/jiafengkevinchen) for the pointer!
- The number of degrees of freedom `k` was computed incorrectly if columns were dropped from the design matrix `X` in the presence of multicollinearity. See [235](https://github.com/s3alfisc/pyfixest/issues/235) for details. Thanks to [jiafengkevinchen](https://github.com/jiafengkevinchen) for the pointer!
- If all variables were dropped due to multicollinearity, an unclear and imprecise error message was produced. See [228](https://github.com/s3alfisc/pyfixest/issues/228) for details. Thanks to [manferdinig](https://github.com/manferdinig) for the pointer!
- If selection `fixef_rm = 'singleton'`, `feols()` and `fepois()` would fail, which has been fixed. [192](https://github.com/s3alfisc/pyfixest/issues/192)

Dependency Requirements:

- For now, sets `formulaic` versions to be `0.6.6` or lower as version `1.0.0` seems to have introduced a problem with the `i()` operator, See [244](https://github.com/s3alfisc/pyfixest/issues/244) for details.
- Drops dependency on `pyhdfe`.

- Significant speedups for CRV1 inference thanks to help by styfenschaer.
- Addition of `fixest` style benchmarks for OLS and Poisson Regression.

Fixes a bug with the separation check for poisson regression 138.

- Fixes bugs with `i(var1, var2)` syntax introduced with `PyFixest` `0.10.10`.